vdsm-patches November 2012

vdsm-patches@lists.fedorahosted.org

50 participants
416 discussions

Change in vdsm[master]: image: unify the prezeroing optimizations
by Federico Simoncelli 19 Oct '15

19 Oct '15

Federico Simoncelli has uploaded a new change for review. Change subject: image: unify the prezeroing optimizations ...................................................................... image: unify the prezeroing optimizations The same prezeroing optimization logic was used in multiple places, this patch unifies it in __optimizedCreateVolume. Change-Id: I0fd90f85e9debf98bcac07d1b8d4b38c319c33f2 Signed-off-by: Federico Simoncelli <fsimonce(a)redhat.com> --- M vdsm/storage/image.py 1 file changed, 43 insertions(+), 45 deletions(-) git pull ssh://gerrit.ovirt.org:29418/vdsm refs/changes/04/8504/1 diff --git a/vdsm/storage/image.py b/vdsm/storage/image.py index e86d94c..19ab078 100644 --- a/vdsm/storage/image.py +++ b/vdsm/storage/image.py @@ -454,6 +454,37 @@ except Exception: self.log.error("Unexpected error", exc_info=True) + def __optimizedCreateVolume(self, domain, imgUUID, size, apparentSize, + volFormat, preallocate, diskType, volUUID, desc, srcImgUUID, + srcVolUUID): + # To avoid 'prezeroing' preallocated volume on NFS domain, + # we create the target volume with minimal size and after + # that we'll change its metadata back to the original size. + if (volFormat == volume.COW_FORMAT + or preallocate == volume.SPARSE_VOL): + volTmpSize = size + else: + volTmpSize = TEMPORARY_VOLUME_SIZE + + domain.createVolume(imgUUID, volTmpSize, volFormat, preallocate, + diskType, volUUID, desc, srcImgUUID, srcVolUUID) + newVolume = domain.produceVolume(imgUUID, volUUID) + + if volFormat == volume.RAW_FORMAT: + extendSize = size + else: + extendSize = apparentSize + + # Extend volume (for LV only) size to the actual size + newVolume.extend((extendSize + 511) / 512) + + # Change destination volume metadata back to the original + # size. Heavy operation, do it only if necessary. + if volTmpSize != size: + newVolume.setSize(size) + + return newVolume + def _createTargetImage(self, destDom, srcSdUUID, imgUUID): # Before actual data copying we need perform several operation # such as: create all volumes, create fake template if needed, ... @@ -500,34 +531,12 @@ # find out src volume parameters volParams = srcVol.getVolumeParams(bs=1) - # To avoid 'prezeroing' preallocated volume on NFS domain, - # we create the target volume with minimal size and after - # that w'll change its metadata back to the original size. - if (volParams['volFormat'] == volume.COW_FORMAT - or volParams['prealloc'] == volume.SPARSE_VOL): - volTmpSize = volParams['size'] - else: - volTmpSize = TEMPORARY_VOLUME_SIZE # in sectors (10M) - - destDom.createVolume(imgUUID=imgUUID, size=volTmpSize, - volFormat=volParams['volFormat'], - preallocate=volParams['prealloc'], - diskType=volParams['disktype'], - volUUID=srcVol.volUUID, - desc=volParams['descr'], - srcImgUUID=pimg, - srcVolUUID=volParams['parent']) - - dstVol = destDom.produceVolume(imgUUID=imgUUID, - volUUID=srcVol.volUUID) - - # Extend volume (for LV only) size to the actual size - dstVol.extend((volParams['apparentsize'] + 511) / 512) - - # Change destination volume metadata back to the original - # size. - if volTmpSize != volParams['size']: - dstVol.setSize(volParams['size']) + dstVol = self.__optimizedCreateVolume( + destDom, imgUUID, volParams['size'], + volParams['apparentsize'], volParams['volFormat'], + volParams['prealloc'], volParams['disktype'], + srcVol.volUUID, volParams['descr'], srcImgUUID=pimg, + srcVolUUID=volParams['parent']) dstChain.append(dstVol) except se.StorageException: @@ -760,25 +769,14 @@ self.log.info("delete image %s on domain %s before overwriting", dstImgUUID, dstSdUUID) self.delete(dstSdUUID, dstImgUUID, postZero, force=True) - # To avoid 'prezeroing' preallocated volume on NFS domain, - # we create the target volume with minimal size and after that w'll change - # its metadata back to the original size. - tmpSize = TEMPORARY_VOLUME_SIZE # in sectors (10M) - destDom.createVolume(imgUUID=dstImgUUID, size=tmpSize, - volFormat=dstVolFormat, preallocate=volParams['prealloc'], - diskType=volParams['disktype'], volUUID=dstVolUUID, desc=descr, - srcImgUUID=volume.BLANK_UUID, srcVolUUID=volume.BLANK_UUID) + dstVol = self.__optimizedCreateVolume( + destDom, dstImgUUID, volParams['size'], + volParams['apparentsize'], dstVolFormat, + volParams['prealloc'], volParams['disktype'], + dstVolUUID, descr, volume.BLANK_UUID, + volume.BLANK_UUID) - dstVol = sdCache.produce(dstSdUUID).produceVolume(imgUUID=dstImgUUID, volUUID=dstVolUUID) - # For convert to 'raw' we need use the virtual disk size instead of apparent size - if dstVolFormat == volume.RAW_FORMAT: - newsize = volParams['size'] - else: - newsize = volParams['apparentsize'] - dstVol.extend(newsize) dstPath = dstVol.getVolumePath() - # Change destination volume metadata back to the original size. - dstVol.setSize(volParams['size']) except se.StorageException, e: self.log.error("Unexpected error", exc_info=True) raise -- To view, visit http://gerrit.ovirt.org/8504 To unsubscribe, visit http://gerrit.ovirt.org/settings Gerrit-MessageType: newchange Gerrit-Change-Id: I0fd90f85e9debf98bcac07d1b8d4b38c319c33f2 Gerrit-PatchSet: 1 Gerrit-Project: vdsm Gerrit-Branch: master Gerrit-Owner: Federico Simoncelli <fsimonce(a)redhat.com>

8 14

Change in vdsm[master]: [wip] sdcache: avoid extra refresh due samplingmethod
by Federico Simoncelli 19 Oct '15

19 Oct '15

Federico Simoncelli has uploaded a new change for review. Change subject: [wip] sdcache: avoid extra refresh due samplingmethod ...................................................................... [wip] sdcache: avoid extra refresh due samplingmethod In order to avoid an extra iscsi rescan (symptomatic of samplingmethod) an additional lock has been introduced to queue the requests when the storage is flagged as stale. Bug-Url: https://bugzilla.redhat.com/show_bug.cgi?id=870768 Change-Id: If178a8eaeb94f1dfe9e0957036dde88f6a22829c Signed-off-by: Federico Simoncelli <fsimonce(a)redhat.com> --- M vdsm/storage/sdc.py 1 file changed, 25 insertions(+), 26 deletions(-) git pull ssh://gerrit.ovirt.org:29418/vdsm refs/changes/74/9274/1 diff --git a/vdsm/storage/sdc.py b/vdsm/storage/sdc.py index f2f4534..978e3fa 100644 --- a/vdsm/storage/sdc.py +++ b/vdsm/storage/sdc.py @@ -62,32 +62,27 @@ STORAGE_UPDATED = 0 STORAGE_STALE = 1 - STORAGE_REFRESHING = 2 def __init__(self, storage_repo): - self._syncroot = threading.Condition() + self._syncDomain = threading.Condition() + self._syncRefresh = threading.Lock() self.__domainCache = {} self.__inProgress = set() self.__staleStatus = self.STORAGE_STALE self.storage_repo = storage_repo def invalidateStorage(self): - with self._syncroot: - self.__staleStatus = self.STORAGE_STALE + self.log.debug("The storages have been invalidated") + self.__staleStatus = self.STORAGE_STALE @misc.samplingmethod def refreshStorage(self): - self.__staleStatus = self.STORAGE_REFRESHING - + # We need to set the __staleStatus value at the beginning because we + # want to keep track of the future invalidateStorage calls that might + # arrive during the rescan procedure. + self.__staleStatus = self.STORAGE_UPDATED multipath.rescan() lvm.invalidateCache() - - # If a new invalidateStorage request came in after the refresh - # started then we cannot flag the storages as updated (force a - # new rescan later). - with self._syncroot: - if self.__staleStatus == self.STORAGE_REFRESHING: - self.__staleStatus = self.STORAGE_UPDATED def produce(self, sdUUID): domain = DomainProxy(self, sdUUID) @@ -98,7 +93,7 @@ return domain def _realProduce(self, sdUUID): - with self._syncroot: + with self._syncDomain: while True: domain = self.__domainCache.get(sdUUID) @@ -109,25 +104,29 @@ self.__inProgress.add(sdUUID) break - self._syncroot.wait() + self._syncDomain.wait() try: - # If multiple calls reach this point and the storage is not - # updated the refreshStorage() sampling method is called - # serializing (and eventually grouping) the requests. - if self.__staleStatus != self.STORAGE_UPDATED: - self.refreshStorage() + # Here we cannot take full advantage of the refreshStorage + # samplingmethod since we might be scheduling an unneeded + # extra rescan. We need an additional lock (_syncRefresh) + # to make sure that __staleStatus is taken in account + # (without affecting all the other external refreshStorage + # calls as it would be if we move this check there). + with self._syncRefresh: + if self.__staleStatus != self.STORAGE_UPDATED: + self.refreshStorage() domain = self._findDomain(sdUUID) - with self._syncroot: + with self._syncDomain: self.__domainCache[sdUUID] = domain return domain finally: - with self._syncroot: + with self._syncDomain: self.__inProgress.remove(sdUUID) - self._syncroot.notifyAll() + self._syncDomain.notifyAll() def _findDomain(self, sdUUID): import blockSD @@ -162,16 +161,16 @@ return uuids def refresh(self): - with self._syncroot: + with self._syncDomain: lvm.invalidateCache() self.__domainCache.clear() def manuallyAddDomain(self, domain): - with self._syncroot: + with self._syncDomain: self.__domainCache[domain.sdUUID] = domain def manuallyRemoveDomain(self, sdUUID): - with self._syncroot: + with self._syncDomain: try: del self.__domainCache[sdUUID] except KeyError: -- To view, visit http://gerrit.ovirt.org/9274 To unsubscribe, visit http://gerrit.ovirt.org/settings Gerrit-MessageType: newchange Gerrit-Change-Id: If178a8eaeb94f1dfe9e0957036dde88f6a22829c Gerrit-PatchSet: 1 Gerrit-Project: vdsm Gerrit-Branch: master Gerrit-Owner: Federico Simoncelli <fsimonce(a)redhat.com>

4 10

Change in vdsm[master]: [WIP] BZ#844656 Release the lock during _findDomain
by Federico Simoncelli 19 Oct '15

19 Oct '15

Federico Simoncelli has uploaded a new change for review. Change subject: [WIP] BZ#844656 Release the lock during _findDomain ...................................................................... [WIP] BZ#844656 Release the lock during _findDomain Signed-off-by: Federico Simoncelli <fsimonce(a)redhat.com> Change-Id: I8088d5fe716a3a08c3e5cef2d2d9a654ee96f60a --- M vdsm/storage/sdc.py 1 file changed, 21 insertions(+), 7 deletions(-) git pull ssh://gerrit.ovirt.org:29418/vdsm refs/changes/22/6822/1 -- To view, visit http://gerrit.ovirt.org/6822 To unsubscribe, visit http://gerrit.ovirt.org/settings Gerrit-MessageType: newchange Gerrit-Change-Id: I8088d5fe716a3a08c3e5cef2d2d9a654ee96f60a Gerrit-PatchSet: 1 Gerrit-Project: vdsm Gerrit-Branch: master Gerrit-Owner: Federico Simoncelli <fsimonce(a)redhat.com>

8 18

Change in vdsm[master]: dump the core of a domain
by shaohef＠linux.vnet.ibm.com 19 Oct '15

19 Oct '15

ShaoHe Feng has uploaded a new change for review. Change subject: dump the core of a domain ...................................................................... dump the core of a domain libvirt support an API to dump the core of a domain on a given file for analysis when guest OS crash. There are two kind of dump files. one is QEMU suspend to disk image. the other is core file which like kdump file butcontains registers' value. It's helpful to support by VDSM to find root cause if a guest gets hang and kdump isn't set up in it. This would be a good RAS feature. Here's the definition of the new API: coreDump: This method will dump the core of a domain on a given file for analysis. Input parameter: vmId - VM UUID to - the core file named by the user flags - defined in libvirt.py VIR_DUMP_CRASH VIR_DUMP_LIVE VIR_DUMP_BYPASS_CACHE VIR_DUMP_RESET VIR_DUMP_MEMORY_ONLY Return value: success: return doneCode failure: return errCode including underlying libvirt error message. Change-Id: If4aac9e747dc7aa64a6ff5ef256a7a4375aa2bb5 Signed-off-by: ShaoHe Feng <shaohef(a)linux.vnet.ibm.com> --- M vdsm/API.py M vdsm/BindingXMLRPC.py M vdsm/define.py M vdsm/libvirtvm.py M vdsm_cli/vdsClient.py 5 files changed, 80 insertions(+), 0 deletions(-) git pull ssh://gerrit.ovirt.org:29418/vdsm refs/changes/29/7329/1 diff --git a/vdsm/API.py b/vdsm/API.py index 19cbb42..e2b24cb 100644 --- a/vdsm/API.py +++ b/vdsm/API.py @@ -244,6 +244,12 @@ self.log.debug("Error creating VM", exc_info=True) return errCode['unexpected'] + def coreDump(self, to, flags): + v = self._cif.vmContainer.get(self._UUID) + if not v: + return errCode['noVM'] + return v.coreDump(to, flags) + def desktopLock(self): """ Lock user session in guest operating system using guest agent. diff --git a/vdsm/BindingXMLRPC.py b/vdsm/BindingXMLRPC.py index cc5300f..be71e6a 100644 --- a/vdsm/BindingXMLRPC.py +++ b/vdsm/BindingXMLRPC.py @@ -208,6 +208,10 @@ vm = API.VM(vmId) return vm.cont() + def vmCoreDump(self, vmId, to, flags): + vm = API.VM(vmId) + return vm.coreDump(to, flags) + def vmReset(self, vmId): vm = API.VM(vmId) return vm.reset() @@ -725,6 +729,7 @@ (self.getVMList, 'list'), (self.vmPause, 'pause'), (self.vmCont, 'cont'), + (self.vmCoreDump, 'coreDump'), (self.vmSnapshot, 'snapshot'), (self.vmMerge, 'merge'), (self.vmMergeStatus, 'mergeStatus'), diff --git a/vdsm/define.py b/vdsm/define.py index 31deb4f..1fedac5 100644 --- a/vdsm/define.py +++ b/vdsm/define.py @@ -114,6 +114,10 @@ 'mergeErr': {'status': {'code': 52, 'message': 'Merge failed'}}, + 'coreDumpErr': {'status': + {'code': 54, + 'message': + 'Failed to get coreDump file'}}, 'recovery': {'status': {'code': 99, 'message': diff --git a/vdsm/libvirtvm.py b/vdsm/libvirtvm.py index 4554fee..cbd9f96 100644 --- a/vdsm/libvirtvm.py +++ b/vdsm/libvirtvm.py @@ -1904,6 +1904,27 @@ self.saveState() + def coreDump(self, to, flags): + + def reportError(key='coreDumpErr', msg=None): + self.log.error("get coreDump failed", exc_info=True) + if msg == None: + error = errCode[key] + else: + error = {'status' : {'code': errCode[key] \ + ['status']['code'], 'message': msg}} + return error + + if self._dom == None: + return reportError() + try: + self._dom.coreDump(to, flags) + except libvirt.libvirtError, e: + if e.get_error_code() == libvirt.VIR_ERR_NO_DOMAIN: + return reportError(key='noVM') + return reportError(msg=e.message) + return {'status': doneCode} + def changeCD(self, drivespec): return self._changeBlockDev('cdrom', 'hdc', drivespec) diff --git a/vdsm_cli/vdsClient.py b/vdsm_cli/vdsClient.py index eeb7c95..cdcd3a8 100644 --- a/vdsm_cli/vdsClient.py +++ b/vdsm_cli/vdsClient.py @@ -1589,6 +1589,33 @@ return status['status']['code'], status['status']['message'] + def coreDump(self, args): + DUMPFLAGS = {'crash': 1 << 0, + 'live': 1 << 1, + 'bypass-cache': 1 << 2, + 'reset': 1 << 3, + 'memory-only': 1 << 4} + flags = 0 + vmId = args[0] + coreFile = args[1] + params = {} + if len(args) > 2: + for arg in args[2:]: + kv = arg.split('=', 1) + if len(kv) < 2: + params[kv[0]] = "True" + else: + params[kv[0]] = kv[1] + for k, v in params.items(): + if v.lower() == "true" or not v: + try: + flags = flags + DUMPFLAGS[k] + except KeyError: + print "unrecognized optoin %s for cormDump command" % k + response = self.s.coreDump(vmId, coreFile, flags) + return response['status']['code'], response['status']['message'] + + if __name__ == '__main__': if _glusterEnabled: serv = ge.GlusterService() @@ -2239,6 +2266,23 @@ ('<vmId> <sdUUID> <imgUUID> <baseVolUUID> <volUUID>', "Take a live snapshot" )), + 'coreDump': (serv.coreDump, + ('<vmId> <file> [live=<True>] ' + '[crash=<True>] [bypass-cache=<True>] ' + '[reset=<True>] [memory-only=<True>]', + "get memeory dump or migration file" + 'optional params:', + 'crash: crash the domain after core dump' + 'default False', + 'live: perform a live core dump if supported, ' + 'default False', + 'bypass-cache: avoid file system cache when saving' + 'default False', + 'reset: reset the domain after core dump' + 'default False', + "memory-only: dump domain's memory only" + 'default False' + )), } if _glusterEnabled: commands.update(ge.getGlusterCmdDict(serv)) -- To view, visit http://gerrit.ovirt.org/7329 To unsubscribe, visit http://gerrit.ovirt.org/settings Gerrit-MessageType: newchange Gerrit-Change-Id: If4aac9e747dc7aa64a6ff5ef256a7a4375aa2bb5 Gerrit-PatchSet: 1 Gerrit-Project: vdsm Gerrit-Branch: master Gerrit-Owner: ShaoHe Feng <shaohef(a)linux.vnet.ibm.com>

16 96

Change in vdsm[master]: [WIP] Added glusterVolumeTop verb
by tjeyasin＠redhat.com 19 Oct '15

19 Oct '15

Hello Ayal Baron, Bala.FA, Saggi Mizrahi, Federico Simoncelli, Dan Kenigsberg, I'd like you to do a code review. Please visit http://gerrit.ovirt.org/7844 to review the following change. Change subject: [WIP] Added glusterVolumeTop verb ...................................................................... [WIP] Added glusterVolumeTop verb Added glusterVolumeTopOpen verb Added glusterVolumeTopRead verb Added glusterVolumeTopWrite verb Added glusterVolumeTopOpenDir verb Added glusterVolumeTopReadDir verb Added glusterVolumeTopReadPerf verb Added glusterVolumeTopWritePerf verb Following is the output structure of glusterVolumeTopOpen {'statusCode' : CODE, 'brickCount': BRICK-COUNT, 'bricks': {BRICK-NAME: {'count':FILE-COUNT, 'currentOpenFds': CURRENT-OPEN-FDS-COUNT, 'maxOpen': MAX-OPEN, 'maxOpenTime': MAX-OPEN-TIME, 'files': [{FILE-NAME: FILE-OPEN-COUNT}, ...] }, ...} } Following is the output structure of glusterVolumeTopRead {'statusCode': CODE, 'brickCount': BRICK-COUNT, 'topOp': TOP-OP, 'bricks': {BRICK-NAME: { 'count': FILE-COUNT, 'files': [{FILE-NAME: FILE-READ-COUNT}, ...]} ,...}} Following is the output structure glusterVolumeTopWrite {'statusCode' : CODE, 'brickCount': BRICK-COUNT, 'topOp': TOP-OP, 'bricks': {BRICK-NAME: {'count': FILE-COUNT, 'files': [{FILE-NAME: FILE-WRITE-COUNT}...]} ,...}} Following is the output structure glusterVolumeTopOpenDir {'statusCode': CODE, 'brickCount': BRICK-COUNT, 'topOp': TOP-OP, 'bricks': {BRICK-NAME: {'count':OPEN-DIR-COUNT, 'files': [{DIR-NAME: DIR-OPEN-COUNT}, ...]} ,...} Following is the output structure glusterVolumeTopReadDir {'statusCode': CODE, 'brickCount': BRICK-COUNT, 'topOp': TOP-OP, 'bricks': {BRICK-NAME: {'count':READ-DIR-COUNT, 'files': [{DIR-NAME: DIR-READ-COUNT}, ...]} ,...} Following is the output structure glusterVolumeTopReadPerf {'statusCode': CODE, 'brickCount': BRICK-COUNT, 'topOp': TOP-OP, 'bricks': {BRICK-NAME: {'fileCount':READ-COUNT, 'throughput': BRICK-WISE-READ-THROUGHPUT, ' timeTaken': TIME-TAKEN, 'files': [{FILE-NAME: {'throughput':FILE-READ-THROUGHPUT, 'time': TIME}}, ...]} ,...}} Following is the output structure glusterVolumeTopWritePerf {'statusCode': CODE, 'brickCount': BRICK-COUNT, 'topOp': TOP-OP, 'bricks': {BRICK-NAME: {'fileCount':WRITE-COUNT, 'throughput': BRICK-WISE-WRITE-THROUGHPUT, ' timeTaken': TIME-TAKEN, 'files': [{FILE-NAME: {'throughput':FILE-WRITE-THROUGHPUT, 'time': TIME}}, ...]} ,...}} Change-Id: I96486363a9acb7472014a67fcd2d5185d4f3c428 Signed-off-by: Timothy Asir <tjeyasin(a)redhat.com> --- M vdsm/gluster/api.py M vdsm/gluster/cli.py M vdsm/gluster/exception.py M vdsm_cli/vdsClientGluster.py 4 files changed, 372 insertions(+), 0 deletions(-) git pull ssh://gerrit.ovirt.org:29418/vdsm refs/changes/44/7844/1 diff --git a/vdsm/gluster/api.py b/vdsm/gluster/api.py index e52430b..3f493e0 100644 --- a/vdsm/gluster/api.py +++ b/vdsm/gluster/api.py @@ -241,6 +241,61 @@ status = self.svdsmProxy.glusterVolumeProfileInfo(volumeName) return {'profileInfo': status} + @exportAsVerb + def volumeTopOpen(self, volumeName, brickName=None, count=None, + options=None): + status = self.svdsmProxy.glusterVolumeTopOpen(volumeName, + brickName, count) + return {'topOpen': status} + + @exportAsVerb + def volumeTopRead(self, volumeName, brickName=None, count=None, + options=None): + status = self.svdsmProxy.glusterVolumeTopRead(volumeName, + brickName, count) + return {'topRead': status} + + @exportAsVerb + def volumeTopWrite(self, volumeName, brickName=None, count=None, + options=None): + status = self.svdsmProxy.glusterVolumeTopWrite(volumeName, + brickName, count) + return {'topWrite': status} + + @exportAsVerb + def volumeTopOpenDir(self, volumeName, brickName=None, count=None, + options=None): + status = self.svdsmProxy.glusterVolumeTopOpenDir(volumeName, + brickName, count) + return {'topOpenDir': status} + + @exportAsVerb + def volumeTopWriteDir(self, volumeName, brickName=None, count=None, + options=None): + status = self.svdsmProxy.glusterVolumeTopWriteDir(volumeName, + brickName, count) + return {'topWriteDir': status} + + @exportAsVerb + def volumeTopReadPerf(self, volumeName, blockSize=None, count=None, + brickName=None, listCount=None, options=None): + status = self.svdsmProxy.glusterVolumeTopReadPerf(volumeName, + blockSize, + count, + brickName, + listCount) + return {'topReadPerf': status} + + @exportAsVerb + def volumeTopWritePerf(self, volumeName, blockSize=None, count=None, + brickName=None, listCount=None, options=None): + status = self.svdsmProxy.glusterVolumeTopWritePerf(volumeName, + blockSize, + count, + brickName, + listCount) + return {'topWritePerf': status} + def getGlusterMethods(gluster): l = [] diff --git a/vdsm/gluster/cli.py b/vdsm/gluster/cli.py index b91a04f..ba4768c 100644 --- a/vdsm/gluster/cli.py +++ b/vdsm/gluster/cli.py @@ -334,6 +334,66 @@ return volumeInfoDict +def _parseGlusterVolumeTopOpen(tree): + bricks = {} + for brick in tree.findall('volTop/brick'): + fileList = [] + for file in brick.findall('file'): + fileList.append({file.find('filename').text: + file.find('count').text}) + bricks[brick.find('name').text] = { + 'count': brick.find('members').text, + 'currentOpen': brick.find('currentOpen').text, + 'maxOpen': brick.find('maxOpen').text, + 'maxOpenTime': brick.find('maxOpenTime').text, + 'files': fileList} + status = { + 'topOp': tree.find('volTop/topOp').text, + 'brickCount': tree.find('volTop/brickCount').text, + 'statusCode': tree.find('opRet').text, + 'bricks': bricks} + return status + + +def _parseGlusterVolumeTop(tree): + bricks = {} + for brick in tree.findall('volTop/brick'): + fileList = [] + for fileTag in brick.findall('file'): + fileList.append({fileTag.find('filename').text: + fileTag.find('count').text}) + bricks[brick.find('name').text] = { + 'count': brick.find('members').text, + 'files': fileList} + status = { + 'topOp': tree.find('volTop/topOp').text, + 'brickCount': tree.find('volTop/brickCount').text, + 'statusCode': tree.find('opRet').text, + 'bricks': bricks} + return status + + +def _parseGlusterVolumeTopPerf(tree): + bricks = {} + for brick in tree.findall('volTop/brick'): + fileList = [] + for fileTag in brick.findall('file'): + fileList.append({fileTag.find('filename').text: + {'count': fileTag.find('count').text, + 'time': fileTag.find('time').text}}) + bricks[brick.find('name').text] = { + 'count': brick.find('members').text, + 'throughput': brick.find('throughput').text, + 'timeTaken': brick.find('timeTaken').text, + 'files': fileList} + status = { + 'topOp': tree.find('volTop/topOp').text, + 'brickCount': tree.find('volTop/brickCount').text, + 'statusCode': tree.find("opRet").text, + 'bricks': bricks} + return status + + def _parseGlusterVolumeProfileInfo(tree): bricks = {} for brick in tree.findall('volProfile/brick'): @@ -819,3 +879,132 @@ return _parseGlusterVolumeProfileInfo(xmltree) except: raise ge.GlusterXmlErrorException(err=out) + + +@exportToSuperVdsm +def volumeTopOpen(volumeName, brickName=None, count=None): + command = _getGlusterVolCmd() + ["top", volumeName, "open"] + if brickName: + command += ["brick", "%s" % brickName] + if count: + command += ["list-cnt", "%s" % count] + try: + xmltree, out = _execGlusterXml(command) + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeTopOpenFailedException(rc=e.rc, err=e.err) + try: + return _parseGlusterVolumeTopOpen(xmltree) + except: + raise ge.GlusterXmlErrorException(err=out) + + +@exportToSuperVdsm +def volumeTopRead(volumeName, brickName=None, count=None): + command = _getGlusterVolCmd() + ["top", volumeName, "read"] + if brickName: + command += ["brick", "%s" % brickName] + if count: + command += ["list-cnt", "%s" % count] + try: + xmltree, out = _execGlusterXml(command) + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeTopReadFailedException(rc=e.rc, err=e.err) + try: + return _parseGlusterVolumeTop(xmltree) + except: + raise ge.GlusterXmlErrorException(err=out) + + +@exportToSuperVdsm +def volumeTopWrite(volumeName, brickName=None, count=None): + command = _getGlusterVolCmd() + ["top", volumeName, "write"] + if brickName: + command += ["brick", "%s" % brickName] + if count: + command += ["list-cnt", "%s" % count] + try: + xmltree, out = _execGlusterXml(command) + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeTopWriteFailedException(rc=e.rc, err=e.err) + try: + return _parseGlusterVolumeTop(xmltree) + except: + raise ge.GlusterXmlErrorException(err=out) + + +@exportToSuperVdsm +def volumeTopOpenDir(volumeName, brickName=None, count=None): + command = _getGlusterVolCmd() + ["top", volumeName, "opendir"] + if brickName: + command += ["brick", "%s" % brickName] + if count: + command += ["list-cnt", "%s" % count] + try: + xmltree, out = _execGlusterXml(command) + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeTopOpenDirFailedException(rc=e.rc, err=e.err) + try: + return _parseGlusterVolumeTop(xmltree) + except: + raise ge.GlusterXmlErrorException(err=out) + + +@exportToSuperVdsm +def volumeTopReadDir(volumeName, bricName=None, count=None): + command = _getGlusterVolCmd() + ["top", volumeName, "write"] + if brickName: + command += ["brick", "%s" % brickName] + if count: + command += ["list-cnt", "%s" % count] + try: + xmltree, out = _execGlusterXml(command) + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeTopReadDirFailedException(rc=e.rc, err=e.err) + try: + return _parseGlusterVolumeTop(xmltree) + except: + raise ge.GlusterXmlErrorException(err=out) + + +@exportToSuperVdsm +def volumeTopReadPerf(volumeName, blockSize=None, count=None, + brickName=None, listCount=None): + command = _getGlusterVolCmd() + ["top", volumeName, "read-perf"] + if blockSize: + command += ["bs", "%s" % blockSize] + if count: + command += ["count", "%s" % count] + if brickName: + command += ["brick", "%s" % brickName] + if listCount: + command += ["list-cnt", "%s" % listCount] + try: + xmltree, out = _execGlusterXml(command) + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeTopReadPerfFailedException(rc=e.rc, err=e.err) + try: + return _parseGlusterVolumeTopPerf(xmltree) + except: + raise ge.GlusterXmlErrorException(err=out) + + +@exportToSuperVdsm +def volumeTopWritePerf(volumeName, blockSize=None, count=None, + brickName=None, listCount=None): + command = _getGlusterVolCmd() + ["top", volumeName, "write-perf"] + if blockSize: + command += ["bs", "%s" % blockSize] + if count: + command += ["count", "%s" % count] + if brickName: + command += ["brick", "%s" % brickName] + if listCount: + command += ["list-cnt", "%s" % listCount] + try: + xmltree, out = _execGlusterXml(command) + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeTopWritePerfFailedException(rc=e.rc, err=e.err) + try: + return _parseGlusterVolumeTopPerf(xmltree) + except: + raise ge.GlusterXmlErrorException(err=out) diff --git a/vdsm/gluster/exception.py b/vdsm/gluster/exception.py index bc20dd0..b392ec8 100644 --- a/vdsm/gluster/exception.py +++ b/vdsm/gluster/exception.py @@ -343,6 +343,41 @@ message = "Volume profile info failed" +class GlusterVolumeTopOpenFailedException(GlusterVolumeException): + code = 4161 + message = "Volume top open failed" + + +class GlusterVolumeTopReadFailedException(GlusterVolumeException): + code = 4162 + message = "Volume top read failed" + + +class GlusterVolumeTopWriteFailedException(GlusterVolumeException): + code = 4163 + message = "Volume top write failed" + + +class GlusterVolumeTopOpenDirFailedException(GlusterVolumeException): + code = 4164 + message = "Volume top open dir failed" + + +class GlusterVolumeTopReadDirFailedException(GlusterVolumeException): + code = 4165 + message = "Volume top read dir failed" + + +class GlusterVolumeTopReadPerfFailedException(GlusterVolumeException): + code = 4166 + message = "Volume top read perf failed" + + +class GlusterVolumeTopWritePerfFailedException(GlusterVolumeException): + code = 4167 + message = "Volume top write perf failed" + + # Host class GlusterHostException(GlusterException): code = 4400 diff --git a/vdsm_cli/vdsClientGluster.py b/vdsm_cli/vdsClientGluster.py index 8422695..3663c63 100644 --- a/vdsm_cli/vdsClientGluster.py +++ b/vdsm_cli/vdsClientGluster.py @@ -221,6 +221,41 @@ pp.pprint(status) return status['status']['code'], status['status']['message'] + def do_glusterVolumeTopOpen(self, args): + status = self.s.glusterVolumeTopOpen(args[0]) + pp.pprint(status) + return status['status']['code'], status['status']['message'] + + def do_glusterVolumeTopRead(self, args): + status = self.s.glusterVolumeTopRead(args[0]) + pp.pprint(status) + return status['status']['code'], status['status']['message'] + + def do_glusterVolumeTopWrite(self, args): + status = self.s.glusterVolumeTopWrite(args[0]) + pp.pprint(status) + return status['status']['code'], status['status']['message'] + + def do_glusterVolumeTopOpenDir(self, args): + status = self.s.glusterVolumeTopOpenDir(args[0]) + pp.pprint(status) + return status['status']['code'], status['status']['message'] + + def do_glusterVolumeTopReadDir(self, args): + status = self.s.glusterVolumeTopReadDir(args[0]) + pp.pprint(status) + return status['status']['code'], status['status']['message'] + + def do_glusterVolumeTopReadPerf(self, args): + status = self.s.glusterVolumeTop(args[0]) + pp.pprint(status) + return status['status']['code'], status['status']['message'] + + def do_glusterVolumeTopWritePerf(self, args): + status = self.s.glusterVolumeTop(args[0]) + pp.pprint(status) + return status['status']['code'], status['status']['message'] + def getGlusterCmdDict(serv): return { @@ -403,4 +438,62 @@ ('<volume_name>\n\t<volume_name> is existing volume name', 'get gluster volume profile info' )), + 'glusterVolumeTopOpen': + (serv.do_glusterVolumeTopOpen, + ('<volume_name> [brick=<existing_brick>] ' + '[count=<list_count>]\n\t' + '<volume_name> is existing volume name\n\t' + 'get volume top open fd count and maximum fd count of ' + 'a given volume with its all brick or specified brick' + )), + 'glusterVolumeTopRead': + (serv.do_glusterVolumeTopRead, + ('<volume_name> [brick=<existing_brick>] ' + '[count=<list_count>]\n\t' + '<volume_name> is existing volume name\n\t' + 'get list of highest read calls on each brick or ' + 'a specified brick of a volume' + )), + 'glusterVolumeTopWrite': + (serv.do_glusterVolumeTopWrite, + ('<volume_name> [brick=<existing_brick>] ' + '[count=<list_count>]\n\t' + '<volume_name> is existing volume name\n\t' + 'get list of highest write calls on each brick or ' + 'a specified brick of a volume' + )), + 'glusterVolumeTopOpenDir': + (serv.do_glusterVolumeTopOpenDir, + ('<volume_name> [brick=<existing_brick>] ' + '[count=<list_count>]\n\t' + '<volume_name> is existing volume name\n\t' + 'get list of highest open calls on directories of each brick ' + 'or a specified brick of a volume' + )), + 'glusterVolumeTopReadDir': + (serv.do_glusterVolumeTopReadDir, + ('<volume_name> [brick=<existing_brick>] ' + '[count=<list_count>]\n\t' + '<volume_name> is existing volume name\n\t' + 'get list of highest read calls on directories of each brick ' + 'or a specified brick of a volume' + )), + 'glusterVolumeTopReadPerf': + (serv.do_glusterVolumeTopReadPerf, + ('<volume_name> [block_size=<block_size>] ' + '[count=<count>] [brick=<existing_brick>] ' + '[list_count=<list_count>]\n\t' + '<volume_name> is existing volume name\n\t' + 'get list of read throughput of files on bricks. ' + 'if the block size and the count is not specified, ' + 'it will give the output based on historical data' + )), + 'glusterVolumeTopWritePerf': + (serv.do_glusterVolumeTopWritePerf, + ('<volume_name> [block_size=<block_size>] ' + '[count=<count>] [brick=<existing_brick>] ' + '[list_count=<list_count>]\n\t' + '<volume_name> is existing volume name\n\t' + 'get list of write throughput of files on bricks' + )), } -- To view, visit http://gerrit.ovirt.org/7844 To unsubscribe, visit http://gerrit.ovirt.org/settings Gerrit-MessageType: newchange Gerrit-Change-Id: I96486363a9acb7472014a67fcd2d5185d4f3c428 Gerrit-PatchSet: 1 Gerrit-Project: vdsm Gerrit-Branch: master Gerrit-Owner: Timothy Asir <tjeyasin(a)redhat.com> Gerrit-Reviewer: Ayal Baron <abaron(a)redhat.com> Gerrit-Reviewer: Bala.FA <barumuga(a)redhat.com> Gerrit-Reviewer: Dan Kenigsberg <danken(a)redhat.com> Gerrit-Reviewer: Federico Simoncelli <fsimonce(a)redhat.com> Gerrit-Reviewer: Saggi Mizrahi <smizrahi(a)redhat.com>

8 25

Change in vdsm[master]: Added gluster tag support in getAllTasks()
by barumuga＠redhat.com 14 Oct '15

14 Oct '15

Hello Ayal Baron, Timothy Asir, Saggi Mizrahi, Federico Simoncelli, Dan Kenigsberg, I'd like you to do a code review. Please visit http://gerrit.ovirt.org/7579 to review the following change. Change subject: Added gluster tag support in getAllTasks() ...................................................................... Added gluster tag support in getAllTasks() If param tag is empty, all tasks including gluster tasks are returned, else tasks those tags are in param tag list are returned. As below verbs are not consumed by engine/RHS-C yet, its OK to differ in compatibility issue now. glusterVolumeRebalanceStart glusterVolumeRebalanceStatus glusterVolumeReplaceBrickStart glusterVolumeReplaceBrickStatus glusterVolumeRemoveBrickStart glusterVolumeRemoveBrickStatus Change-Id: I9c765cbfebb5ba22f0d21efa04c824ea4daf6432 Signed-off-by: Bala.FA <barumuga(a)redhat.com> --- M tests/gluster_cli_tests.py M vdsm/gluster/cli.py M vdsm/gluster/exception.py M vdsm/storage/taskManager.py 4 files changed, 367 insertions(+), 95 deletions(-) git pull ssh://gerrit.ovirt.org:29418/vdsm refs/changes/79/7579/1 diff --git a/tests/gluster_cli_tests.py b/tests/gluster_cli_tests.py index f442893..9c6357c 100644 --- a/tests/gluster_cli_tests.py +++ b/tests/gluster_cli_tests.py @@ -28,6 +28,7 @@ from gluster import cli as gcli except ImportError: pass +import xml.etree.cElementTree as etree class GlusterCliTests(TestCaseBase): @@ -115,3 +116,74 @@ def test_parsePeerStatus(self): self._parsePeerStatus_empty_test() self._parsePeerStatus_test() + + def _parseVolumeStatusAll_test(self): + out = """<?xml version="1.0" encoding="UTF-8" standalone="yes"?> +<cliOutput> + <opRet>0</opRet> + <opErrno>0</opErrno> + <opErrstr></opErrstr> + <volumes> + <volume> + <name>V1</name> + <id>03eace73-9197-49d0-a877-831bc6e9dac2</id> + <tasks> + <task> + <name>rebalance</name> + <id>12345473-9197-49d0-a877-831bc6e9dac2</id> + </task> + </tasks> + </volume> + <volume> + <name>V2</name> + <id>03eace73-1237-49d0-a877-831bc6e9dac2</id> + <tasks> + <task> + <name>replace-brick</name> + <id>12345473-1237-49d0-a877-831bc6e9dac2</id> + <sourceBrick>192.168.122.167:/tmp/V2-b1</sourceBrick> + <destBrick>192.168.122.168:/tmp/V2-b1</destBrick> + </task> + </tasks> + </volume> + <volume> + <name>V3</name> + <id>03eace73-1237-1230-a877-831bc6e9dac2</id> + <tasks> + <task> + <name>remove-brick</name> + <id>12345473-1237-1230-a877-831bc6e9dac2</id> + <BrickCount>2</BrickCount> + <brick>192.168.122.167:/tmp/V3-b1</brick> + <brick>192.168.122.168:/tmp/V3-b1</brick> + </task> + </tasks> + </volume> + </volumes> +</cliOutput>""" + tree = etree.fromstring(out) + status = gcli._parseVolumeStatusAll(tree) + self.assertEquals(status, + {'12345473-1237-1230-a877-831bc6e9dac2': + {'bricks': ['192.168.122.167:/tmp/V3-b1', + '192.168.122.168:/tmp/V3-b1'], + 'taskType': 'remove-brick', + 'volumeId': + '03eace73-1237-1230-a877-831bc6e9dac2', + 'volumeName': 'V3'}, + '12345473-1237-49d0-a877-831bc6e9dac2': + {'bricks': ['192.168.122.167:/tmp/V2-b1', + '192.168.122.168:/tmp/V2-b1'], + 'taskType': 'replace-brick', + 'volumeId': + '03eace73-1237-49d0-a877-831bc6e9dac2', + 'volumeName': 'V2'}, + '12345473-9197-49d0-a877-831bc6e9dac2': + {'bricks': [], + 'taskType': 'rebalance', + 'volumeId': + '03eace73-9197-49d0-a877-831bc6e9dac2', + 'volumeName': 'V1'}}) + + def test_parseVolumeStatusAll(self): + self._parseVolumeStatusAll_test() diff --git a/vdsm/gluster/cli.py b/vdsm/gluster/cli.py index 95de106..1f464f6 100644 --- a/vdsm/gluster/cli.py +++ b/vdsm/gluster/cli.py @@ -84,6 +84,55 @@ raise ge.GlusterCmdFailedException(rc=rv, err=[msg]) +class TaskType: + REBALANCE = 'rebalance' + REPLACE_BRICK = 'replace-brick' + REMOVE_BRICK = 'remove-brick' + + +def _parseVolumeStatusAll(tree): + """ + returns {TaskId: {'volumeName': VolumeName, + 'volumeId': VolumeId, + 'taskType': TaskType, + 'bricks': BrickList}, ...} + """ + tasks = {} + for el in tree.findall('volumes/volume'): + volumeName = el.find('name').text + volumeId = el.find('id').text + for c in el.findall('tasks/task'): + taskType = c.find('name').text + taskId = c.find('id').text + bricks = [] + if taskType == TaskType.REPLACE_BRICK: + bricks.append(c.find('sourceBrick').text) + bricks.append(c.find('destBrick').text) + elif taskType == TaskType.REMOVE_BRICK: + for b in c.findall('brick'): + bricks.append(b.text) + elif taskType == TaskType.REBALANCE: + pass + tasks[taskId] = {'volumeName': volumeName, + 'volumeId': volumeId, + 'taskType': taskType, + 'bricks': bricks} + return tasks + + +@exportToSuperVdsm +def volumeStatusAll(): + command = _getGlusterVolCmd() + ["status", "all"] + try: + xmltree, out = _execGlusterXml(command) + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeStatusAllFailedException(rc=e.rc, err=e.err) + try: + return _parseVolumeStatusAll(xmltree) + except: + raise ge.GlusterXmlErrorException(err=out) + + def _parseVolumeInfo(out): if not out[0].strip(): del out[0] @@ -300,11 +349,15 @@ command.append("start") if force: command.append("force") - rc, out, err = _execGluster(command) - if rc: - raise ge.GlusterVolumeRebalanceStartFailedException(rc, out, err) - else: - return True + try: + xmltree, out = _execGlusterXml(command) + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeRebalanceStartFailedException(rc=e.rc, + err=e.err) + try: + return {'taskId': xmltree.find('id').text} + except: + raise ge.GlusterXmlErrorException(err=out) @exportToSuperVdsm @@ -312,84 +365,147 @@ command = _getGlusterVolCmd() + ["rebalance", volumeName, "stop"] if force: command.append('force') - rc, out, err = _execGluster(command) - if rc: - raise ge.GlusterVolumeRebalanceStopFailedException(rc, out, err) - else: + try: + _execGlusterXml(command) return True + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeRebalanceStopFailedException(rc=e.rc, + err=e.err) + + +class TaskStatus(): + RUNNING = 'RUNNING' + FAILED = 'FAILED' + COMPLETED = 'COMPLETED' + + +def _parseVolumeRebalanceRemoveBrickStatus(xmltree, mode): + """ + returns {'taskId': UUID, + 'host': [{'name': NAME, + 'id': HOSTID, + 'filesScanned': INT, + 'filesMoved': INT, + 'filesFailed': INT, + 'totalSizeMoved': INT, + 'status': TaskStatus},...] + 'summary': {'filesScanned': INT, + 'filesMoved': INT, + 'filesFailed': INT, + 'totalSizeMoved': INT, + 'status': TaskStatus}} + """ + if mode == 'rebalance': + tree = xmltree.find('volRebalance') + elif mode == 'remove-brick': + tree = xmltree.find('volRemoveBrick') + else: + return + status = \ + {'taskId': tree.find('id').text, + 'summary': \ + {'filesScanned': int(tree.find('summary/filesScanned').text), + 'filesMoved': int(tree.find('summary/filesMoved').text), + 'filesFailed': int(tree.find('summary/filesFailed').text), + 'totalSizeMoved': int(tree.find('summary/totalSizeMoved').text), + 'status': tree.find('summary/status').text}, + 'host': []} + for el in tree.findall('node'): + status['host'].append({'name': el.find('name').text, + 'id': el.find('id').text, + 'filesScanned': + int(el.find('filesScanned').text), + 'filesMoved': int(el.find('filesMoved').text), + 'filesFailed': int(el.find('filesFailed').text), + 'totalSizeMoved': + int(el.find('totalSizeMoved').text), + 'status': el.find('status').text}) + return status + + +def _parseVolumeRebalanceStatus(tree): + return _parseVolumeRebalanceRemoveBrickStatus(tree, 'rebalance') @exportToSuperVdsm def volumeRebalanceStatus(volumeName): - rc, out, err = _execGluster(_getGlusterVolCmd() + ["rebalance", volumeName, - "status"]) - if rc: - raise ge.GlusterVolumeRebalanceStatusFailedException(rc, out, err) - if 'in progress' in out[0]: - return BrickStatus.RUNNING, "\n".join(out) - elif 'complete' in out[0]: - return BrickStatus.COMPLETED, "\n".join(out) - else: - return BrickStatus.UNKNOWN, "\n".join(out) + command = _getGlusterVolCmd() + ["rebalance", volumeName, "status"] + try: + xmltree, out = _execGlusterXml(command) + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeRebalanceStatusFailedException(rc=e.rc, + err=e.err) + try: + return _parseVolumeRebalanceStatus(xmltree) + except: + raise ge.GlusterXmlErrorException(err=out) @exportToSuperVdsm def volumeReplaceBrickStart(volumeName, existingBrick, newBrick): - rc, out, err = _execGluster(_getGlusterVolCmd() + ["replace-brick", - volumeName, - existingBrick, newBrick, - "start"]) - if rc: - raise ge.GlusterVolumeReplaceBrickStartFailedException(rc, out, err) - else: - return True + command = _getGlusterVolCmd() + ["replace-brick", volumeName, + existingBrick, newBrick, "start"] + try: + xmltree, out = _execGlusterXml(command) + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeReplaceBrickStartFailedException(rc=e.rc, + err=e.err) + try: + return {'taskId': xmltree.find('id').text} + except: + raise ge.GlusterXmlErrorException(err=out) @exportToSuperVdsm def volumeReplaceBrickAbort(volumeName, existingBrick, newBrick): - rc, out, err = _execGluster(_getGlusterVolCmd() + ["replace-brick", - volumeName, - existingBrick, newBrick, - "abort"]) - if rc: - raise ge.GlusterVolumeReplaceBrickAbortFailedException(rc, out, err) - else: + command = _getGlusterVolCmd() + ["replace-brick", volumeName, + existingBrick, newBrick, "abort"] + try: + _execGlusterXml(command) return True + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeReplaceBrickAbortFailedException(rc=e.rc, + err=e.err) @exportToSuperVdsm def volumeReplaceBrickPause(volumeName, existingBrick, newBrick): - rc, out, err = _execGluster(_getGlusterVolCmd() + ["replace-brick", - volumeName, - existingBrick, newBrick, - "pause"]) - if rc: - raise ge.GlusterVolumeReplaceBrickPauseFailedException(rc, out, err) - else: + command = _getGlusterVolCmd() + ["replace-brick", volumeName, + existingBrick, newBrick, "pause"] + try: + _execGlusterXml(command) return True + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeReplaceBrickPauseFailedException(rc=e.rc, + err=e.err) + + +def _parseVolumeReplaceBrickStatus(tree): + """ + returns {'taskId': UUID, + 'filesMoved': INT, + 'movingFile': STRING, + 'status': TaskStatus}} + """ + return {'taskId': tree.find('volReplaceBrick/id').text, + 'filesMoved': int(tree.find('volReplaceBrick/filesMoved').text), + 'movingFile': tree.find('volReplaceBrick/movingFile').text, + 'status': tree.find('volReplaceBrick/status').text} @exportToSuperVdsm def volumeReplaceBrickStatus(volumeName, existingBrick, newBrick): - rc, out, err = _execGluster(_getGlusterVolCmd() + ["replace-brick", - volumeName, - existingBrick, newBrick, - "status"]) - if rc: - raise ge.GlusterVolumeReplaceBrickStatusFailedException(rc, out, - err) - message = "\n".join(out) - statLine = out[0].strip().upper() - if BrickStatus.PAUSED in statLine: - return BrickStatus.PAUSED, message - elif statLine.endswith('MIGRATION COMPLETE'): - return BrickStatus.COMPLETED, message - elif statLine.startswith('NUMBER OF FILES MIGRATED'): - return BrickStatus.RUNNING, message - elif statLine.endswith("UNKNOWN"): - return BrickStatus.UNKNOWN, message - else: - return BrickStatus.NA, message + command = _getGlusterVolCmd() + ["replace-brick", volumeName, + existingBrick, newBrick, "status"] + try: + xmltree, out = _execGlusterXml(command) + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeReplaceBrickStatusFailedException(rc=e.rc, + err=e.err) + try: + return _parseVolumeReplaceBrickStatus(xmltree) + except: + raise ge.GlusterXmlErrorException(err=out) @exportToSuperVdsm @@ -399,12 +515,12 @@ existingBrick, newBrick, "commit"] if force: command.append('force') - rc, out, err = _execGluster(command) - if rc: - raise ge.GlusterVolumeReplaceBrickCommitFailedException(rc, out, - err) - else: + try: + _execGlusterXml(command) return True + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeReplaceBrickCommitFailedException(rc=e.rc, + err=e.err) @exportToSuperVdsm @@ -413,12 +529,15 @@ if replicaCount: command += ["replica", "%s" % replicaCount] command += brickList + ["start"] - - rc, out, err = _execGluster(command) - if rc: - raise ge.GlusterVolumeRemoveBrickStartFailedException(rc, out, err) - else: - return True + try: + xmltree, out = _execGlusterXml(command) + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeRemoveBrickStartFailedException(rc=e.rc, + err=e.err) + try: + return {'taskId': xmltree.find('id').text} + except: + raise ge.GlusterXmlErrorException(err=out) @exportToSuperVdsm @@ -427,12 +546,16 @@ if replicaCount: command += ["replica", "%s" % replicaCount] command += brickList + ["stop"] - rc, out, err = _execGluster(command) - - if rc: - raise ge.GlusterVolumeRemoveBrickStopFailedException(rc, out, err) - else: + try: + _execGlusterXml(command) return True + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeRemoveBrickStopFailedException(rc=e.rc, + err=e.err) + + +def _parseVolumeRemoveBrickStatus(tree): + return _parseVolumeRebalanceRemoveBrickStatus(tree, 'remove-brick') @exportToSuperVdsm @@ -441,12 +564,15 @@ if replicaCount: command += ["replica", "%s" % replicaCount] command += brickList + ["status"] - rc, out, err = _execGluster(command) - - if rc: - raise ge.GlusterVolumeRemoveBrickStatusFailedException(rc, out, err) - else: - return "\n".join(out) + try: + xmltree, out = _execGlusterXml(command) + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeRemoveBrickStatusFailedException(rc=e.rc, + err=e.err) + try: + return _parseVolumeRemoveBrickStatus(xmltree) + except: + raise ge.GlusterXmlErrorException(err=out) @exportToSuperVdsm @@ -455,12 +581,12 @@ if replicaCount: command += ["replica", "%s" % replicaCount] command += brickList + ["commit"] - rc, out, err = _execGluster(command) - - if rc: - raise ge.GlusterVolumeRemoveBrickCommitFailedException(rc, out, err) - else: + try: + _execGlusterXml(command) return True + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeRemoveBrickCommitFailedException(rc=e.rc, + err=e.err) @exportToSuperVdsm @@ -469,12 +595,12 @@ if replicaCount: command += ["replica", "%s" % replicaCount] command += brickList + ["force"] - rc, out, err = _execGluster(command) - - if rc: - raise ge.GlusterVolumeRemoveBrickForceFailedException(rc, out, err) - else: + try: + _execGlusterXml(command) return True + except ge.GlusterCmdFailedException, e: + raise ge.GlusterVolumeRemoveBrickForceFailedException(rc=e.rc, + err=e.err) @exportToSuperVdsm diff --git a/vdsm/gluster/exception.py b/vdsm/gluster/exception.py index f4f497b..f209885 100644 --- a/vdsm/gluster/exception.py +++ b/vdsm/gluster/exception.py @@ -323,6 +323,11 @@ message = "Volume remove brick force failed" +class GlusterVolumeStatusAllFailedException(GlusterVolumeException): + code = 4158 + message = "Volume status all failed" + + # Host class GlusterHostException(GlusterException): code = 4400 diff --git a/vdsm/storage/taskManager.py b/vdsm/storage/taskManager.py index 3bc12f3..0a269cd 100644 --- a/vdsm/storage/taskManager.py +++ b/vdsm/storage/taskManager.py @@ -25,6 +25,12 @@ import storage_exception as se from task import Task, Job, TaskCleanType from threadPool import ThreadPool +try: + from gluster import cli as gcli + from gluster import exception as ge + _glusterEnabled = True +except ImportError: + _glusterEnabled = False class TaskManager: @@ -113,19 +119,82 @@ self.log.debug("Return: %s", subRes) return subRes - def getAllTasks(self): + def _getAllGlusterTasks(self): """ - Return Tasks for all public tasks. + Return all gluster tasks + """ + subRes = {} + if not _glusterEnabled: + return subRes + + for taskId, value in gcli.volumeStatusAll(): + msg = '' + state = '' + try: + if value['taskType'] == gcli.TaskType.REBALANCE: + status = gcli.volumeRebalanceStatus(value['volumeName']) + msg = ('Files [scanned: %d, moved: %d, failed: %d], ' + 'Total size moved: %d') % \ + (status['summary']['filesScanned'], + status['summary']['filesMoved'], + status['summary']['filesFailed'], + status['summary']['totalSizeMoved']) + state = status['summary']['status'] + elif value['taskType'] == gcli.TaskType.REMOVE_BRICK: + status = gcli.volumeRemoveBrickStatus(value['volumeName'], + value['bricks']) + msg = ('Files [scanned: %d, moved: %d, failed: %d], ' + 'Total size moved: %d') % \ + (status['summary']['filesScanned'], + status['summary']['filesMoved'], + status['summary']['filesFailed'], + status['summary']['totalSizeMoved']) + state = status['summary']['status'] + elif value['taskType'] == gcli.TaskType.REPLACE_BRICK: + status = gcli.volumeReplaceBrickStatus(value['volumeName'], + value['bricks'][0], + value['bricks'][1]) + msg = 'Files moved: %d, Moving file: %s' % \ + (status['filesMoved'], status['movingFile']) + state = status['status'] + except ge.GlusterException: + self.log.error("gluster exception occured", exc_info=True) + + subRes[taskId] = {"id": taskId, + "verb": value['volumeName'], + "state": state, + "code": value['taskType'], + "message": msg, + "result": '', + "tag": 'gluster'} + return subRes + + def getAllTasks(self, tag=[]): + """ + Return Tasks for all public tasks if param tag is empty, + else return tasks those tags are in param tag. """ self.log.debug("Entry.") subRes = {} for taskID, task in self._tasks.items(): try: - subRes[taskID] = task.getDetails() + if not tag: + subRes[taskID] = task.getDetails() + elif task.getTags() in tag: + subRes[taskID] = task.getDetails() except se.UnknownTask: # Return info for existing tasks only. self.log.warn("Unknown task %s. Maybe task was already " "cleared.", taskID) + + try: + if not tag: + subRes.update(self._getAllGlusterTasks()) + elif 'gluster' in tag: + subRes.update(self._getAllGlusterTasks()) + except ge.GlusterException: + self.log.error("gluster exception occured", exc_info=True) + self.log.debug("Return: %s", subRes) return subRes -- To view, visit http://gerrit.ovirt.org/7579 To unsubscribe, visit http://gerrit.ovirt.org/settings Gerrit-MessageType: newchange Gerrit-Change-Id: I9c765cbfebb5ba22f0d21efa04c824ea4daf6432 Gerrit-PatchSet: 1 Gerrit-Project: vdsm Gerrit-Branch: master Gerrit-Owner: Bala.FA <barumuga(a)redhat.com> Gerrit-Reviewer: Ayal Baron <abaron(a)redhat.com> Gerrit-Reviewer: Dan Kenigsberg <danken(a)redhat.com> Gerrit-Reviewer: Federico Simoncelli <fsimonce(a)redhat.com> Gerrit-Reviewer: Saggi Mizrahi <smizrahi(a)redhat.com> Gerrit-Reviewer: Timothy Asir <tjeyasin(a)redhat.com>

5 16

Change in vdsm[master]: add xmlrpcTests for cpu pinning
by lvroyce＠linux.vnet.ibm.com 15 Sep '14

15 Sep '14

Royce Lv has uploaded a new change for review. Change subject: add xmlrpcTests for cpu pinning ...................................................................... add xmlrpcTests for cpu pinning Change-Id: Ia865f0d5eb4c9aabff6cef57b088c55df73a309e Signed-off-by: Royce Lv<lvroyce(a)linux.vnet.ibm.com> --- M tests/functional/xmlrpcTests.py M tests/vdsClientTests.py 2 files changed, 40 insertions(+), 0 deletions(-) git pull ssh://gerrit.ovirt.org:29418/vdsm refs/changes/12/8412/1 diff --git a/tests/functional/xmlrpcTests.py b/tests/functional/xmlrpcTests.py index 9c865db..2684d0f 100644 --- a/tests/functional/xmlrpcTests.py +++ b/tests/functional/xmlrpcTests.py @@ -174,3 +174,33 @@ destroyResult = self.s.destroy(VMID) self.assertVdsOK(destroyResult) + + def testCpuPin(self): + self.skipNoKVM() + + def assertVMAndGuestUp(): + self.assertVmUp(VMID) + self.assertGuestUp(VMID) + + VMID = '77777777-ffff-3333-aaaa-222222222222' + + with kernelBootImages() as (kernelPath, initramfsPath): + conf = {'display': 'vnc', + 'kernel': kernelPath, + 'initrd': initramfsPath, + 'kernelArgs': 'rd.break=cmdline rd.shell rd.skipfsck', + 'kvmEnable': 'true', + 'memSize': '256', + 'vmId': VMID, + 'vmName': 'vdsm_testPinVM', + 'vmType': 'kvm', + 'cpuPinning': {'emulator': '0', '0': '1'}} + + try: + self.assertVdsOK(self.s.create(conf)) + # wait 65 seconds for VM to come up until timeout + self.retryAssert(assertVMAndGuestUp, 65, 1) + finally: + destroyResult = self.s.destroy(VMID) + + self.assertVdsOK(destroyResult) diff --git a/tests/vdsClientTests.py b/tests/vdsClientTests.py index abf3242..57e6e74 100644 --- a/tests/vdsClientTests.py +++ b/tests/vdsClientTests.py @@ -118,3 +118,13 @@ allArgs[-1] = 'cpuPinning={0:1,1:0}' r4 = serv.do_create(['/dev/null'] + allArgs) self.assertNotEquals(r4, expectResult) + + # test just pin emulator + allArgs[-1] = "cpuPinning={emulator:1-3}" + r5 = serv.do_create(['/dev/null'] + allArgs) + self.assertEquals(r5['cpuPinning'],{'emulator':'1-3'}) + + # test pin emultor and vcpu + allArgs[-1] = "cpuPinning={emulator:1-3,1:0}" + r6 = serv.do_create(['/dev/null'] + allArgs) + self.assertEquals(r6['cpuPinning'],{'emulator':'1-3','1':'0'}) -- To view, visit http://gerrit.ovirt.org/8412 To unsubscribe, visit http://gerrit.ovirt.org/settings Gerrit-MessageType: newchange Gerrit-Change-Id: Ia865f0d5eb4c9aabff6cef57b088c55df73a309e Gerrit-PatchSet: 1 Gerrit-Project: vdsm Gerrit-Branch: master Gerrit-Owner: Royce Lv <lvroyce(a)linux.vnet.ibm.com>

10 39

Change in vdsm[master]: [WIP] Towards a more (block) secure HSM.
by ewarszaw＠redhat.com 21 Jul '14

21 Jul '14

Eduardo has uploaded a new change for review. Change subject: [WIP] Towards a more (block) secure HSM. ...................................................................... [WIP] Towards a more (block) secure HSM. Change-Id: I30df4ee5cdb6b44cf14d8cb155436aac7442a07d --- M vdsm/storage/hsm.py M vdsm/storage/lvm.py M vdsm/storage/sp.py 3 files changed, 25 insertions(+), 5 deletions(-) git pull ssh://gerrit.ovirt.org:29418/vdsm refs/changes/18/2218/1 -- To view, visit http://gerrit.ovirt.org/2218 To unsubscribe, visit http://gerrit.ovirt.org/settings Gerrit-MessageType: newchange Gerrit-Change-Id: I30df4ee5cdb6b44cf14d8cb155436aac7442a07d Gerrit-PatchSet: 1 Gerrit-Project: vdsm Gerrit-Branch: master Gerrit-Owner: Eduardo <ewarszaw(a)redhat.com>

6 28

Change in vdsm[master]: [WIP] Implement a process to do dangerous IO in C
by smizrahi＠redhat.com 23 Jun '14

23 Jun '14

Saggi Mizrahi has uploaded a new change for review. Change subject: [WIP] Implement a process to do dangerous IO in C ...................................................................... [WIP] Implement a process to do dangerous IO in C This replaces the process pool with a process that can serve multiple requests written in C. This implementation is much more scalable and lightweight. Should solve bugs related to running out of helpers, logging getting suck, python forking deadlocking, running out of memory and other things as well. The communication between VDSM and the IOProcess is done with json objects. The IOProcess starts with 3 thread: 1. requestReader - reads requests from the pipe, builds a DOM representation of it and queues it up for handling 2. responseWriter - gets response DOMs from the queue converts them to a JSON string and send it over the pipe 3. requestHandler - pops requests from the queue and provisions threads for handling them. Currently we I just allocate a new thread per request. If there is ever a need to have a thread pool this is where the load balancing is going to sit. Each request gets the are as a JsonNode and returns a response that is a JsonNode as well. Most exported functions are pretty trivial and are a good example on how to write new ones. Unlink the ProcessPoolHelper, high level commands sit of the OopWrapper and are run from the client side instead of being implemented in C on the IOProcess side. Change-Id: Ie4664d5330debbe38ba33b74ebb586ac42913b4a Signed-off-by: Saggi Mizrahi <smizrahi(a)redhat.com> --- M configure.ac M tests/Makefile.am A tests/ioprocessTests.py A tests/outOfProcessTests.py D tests/processPoolTests.py M vdsm.spec.in M vdsm/constants.py.in M vdsm/storage/Makefile.am M vdsm/storage/fileSD.py M vdsm/storage/fileUtils.py M vdsm/storage/fileVolume.py A vdsm/storage/ioprocess.py A vdsm/storage/ioprocess/.gitignore A vdsm/storage/ioprocess/Makefile.am A vdsm/storage/ioprocess/exported-functions.c A vdsm/storage/ioprocess/exported-functions.h A vdsm/storage/ioprocess/ioprocess.c A vdsm/storage/ioprocess/json-dom-generator.c A vdsm/storage/ioprocess/json-dom-generator.h A vdsm/storage/ioprocess/json-dom-parser.c A vdsm/storage/ioprocess/json-dom-parser.h A vdsm/storage/ioprocess/json-dom.c A vdsm/storage/ioprocess/json-dom.h M vdsm/storage/misc.py M vdsm/storage/nfsSD.py M vdsm/storage/outOfProcess.py D vdsm/storage/processPool.py M vdsm/storage/sd.py M vdsm/storage/sp.py M vdsm/storage/task.py 30 files changed, 3,018 insertions(+), 666 deletions(-) git pull ssh://gerrit.ovirt.org:29418/vdsm refs/changes/46/3946/1 -- To view, visit http://gerrit.ovirt.org/3946 To unsubscribe, visit http://gerrit.ovirt.org/settings Gerrit-MessageType: newchange Gerrit-Change-Id: Ie4664d5330debbe38ba33b74ebb586ac42913b4a Gerrit-PatchSet: 1 Gerrit-Project: vdsm Gerrit-Branch: master Gerrit-Owner: Saggi Mizrahi <smizrahi(a)redhat.com>

4 11

Change in vdsm[master]: Move fencing logic out of API.py
by smizrahi＠redhat.com 23 Jun '14

23 Jun '14

Saggi Mizrahi has uploaded a new change for review. Change subject: Move fencing logic out of API.py ...................................................................... Move fencing logic out of API.py API.py should not contain logic. In this patch I also: - Separate the `status` action to a different function because it does something completely different from the other action and has different logic. - Refactor for better code reusability readability. - Use real exception internally and have the API layer translate them to actual exceptions. - Don't support port strings in methods, just in the interface. - Don't support string booleans in the methods, just in the interface. - Added shutdownEvent to ClientIF so there is a public standard way of inspecting vdsm shutdown. - Make the fencing logic use betterPopen Change-Id: I944c6548a42612f705a410fb4290215451bca035 Signed-off-by: Saggi Mizrahi <smizrahi(a)redhat.com> --- M vdsm.spec.in M vdsm/API.py M vdsm/Makefile.am M vdsm/clientIF.py 4 files changed, 25 insertions(+), 72 deletions(-) git pull ssh://gerrit.ovirt.org:29418/vdsm refs/changes/90/7190/1 diff --git a/vdsm.spec.in b/vdsm.spec.in index 23afd6b..f122cb5 100644 --- a/vdsm.spec.in +++ b/vdsm.spec.in @@ -558,6 +558,7 @@ %{_datadir}/%{vdsm_name}/caps.py* %{_datadir}/%{vdsm_name}/clientIF.py* %{_datadir}/%{vdsm_name}/API.py* +%{_datadir}/%{vdsm_name}/fenceAgent.py* %{_datadir}/%{vdsm_name}/hooking.py* %{_datadir}/%{vdsm_name}/hooks.py* %{_datadir}/%{vdsm_name}/libvirtev.py* diff --git a/vdsm/API.py b/vdsm/API.py index ae04e01..5826d81 100644 --- a/vdsm/API.py +++ b/vdsm/API.py @@ -19,12 +19,9 @@ # import os -import signal import copy -import subprocess import pickle import time -import threading import logging from vdsm import utils @@ -40,6 +37,7 @@ from vdsm.define import doneCode, errCode, Kbytes, Mbytes import caps from vdsm.config import config +import fenceAgent import supervdsm @@ -981,82 +979,32 @@ agent is one of (rsa, ilo, drac5, ipmilan, etc) action can be one of (status, on, off, reboot).""" - def waitForPid(p, inp): - """ Wait until p.pid exits. Kill it if vdsm exists before. """ - try: - p.stdin.write(inp) - p.stdin.close() - while p.poll() is None: - if not self._cif._enabled: - self.log.debug('killing fence script pid %s', p.pid) - os.kill(p.pid, signal.SIGTERM) - time.sleep(1) - try: - # improbable race: p.pid may now belong to another - # process - os.kill(p.pid, signal.SIGKILL) - except: - pass - return - time.sleep(1) - self.log.debug('rc %s inp %s out %s err %s', p.returncode, - hidePasswd(inp), - p.stdout.read(), p.stderr.read()) - except: - self.log.error("Error killing fence script", exc_info=True) - - def hidePasswd(text): - cleantext = '' - for line in text.splitlines(True): - if line.startswith('passwd='): - line = 'passwd=XXXX\n' - cleantext += line - return cleantext - self.log.debug('fenceNode(addr=%s,port=%s,agent=%s,user=%s,' + 'passwd=%s,action=%s,secure=%s,options=%s)', addr, port, agent, username, 'XXXX', action, secure, options) - if action not in ('status', 'on', 'off', 'reboot'): - raise ValueError('illegal action ' + action) + secure = utils.tobool(secure) + port = int(port) - script = constants.EXT_FENCE_PREFIX + agent + if action == "status": + try: + power = fenceAgent.getFenceStatus(addr, port, agent, username, + password, secure, options) + + return {'status': doneCode, + 'power': power} + + except fenceAgent.FenceStatusCheckError as e: + return {'status': {'code': 1, 'message': str(e)}} try: - p = subprocess.Popen([script], stdin=subprocess.PIPE, - stdout=subprocess.PIPE, stderr=subprocess.PIPE, - close_fds=True) - except OSError, e: - if e.errno == os.errno.ENOENT: - return errCode['fenceAgent'] - raise + fenceAgent.fenceNode(addr, port, agent, username, password, secure, + options, self._cif.shutdownEvent) - inp = ('agent=fence_%s\nipaddr=%s\nlogin=%s\noption=%s\n' + - 'passwd=%s\n') % (agent, addr, username, action, password) - if port != '': - inp += 'port=%s\n' % (port,) - if utils.tobool(secure): - inp += 'secure=yes\n' - inp += options - if action == 'status': - out, err = p.communicate(inp) - self.log.debug('rc %s in %s out %s err %s', p.returncode, - hidePasswd(inp), out, err) - if not 0 <= p.returncode <= 2: - return {'status': {'code': 1, - 'message': out + err}} - message = doneCode['message'] - if p.returncode == 0: - power = 'on' - elif p.returncode == 2: - power = 'off' - else: - power = 'unknown' - message = out + err - return {'status': {'code': 0, 'message': message}, - 'power': power} - threading.Thread(target=waitForPid, args=(p, inp)).start() - return {'status': doneCode} + return {'status': doneCode} + + except fenceAgent.UnsupportedFencingAgentError: + return errCode['fenceAgent'] def ping(self): "Ping the server. Useful for tests" diff --git a/vdsm/Makefile.am b/vdsm/Makefile.am index 62ba982..4a7bc57 100644 --- a/vdsm/Makefile.am +++ b/vdsm/Makefile.am @@ -32,6 +32,7 @@ clientIF.py \ configNetwork.py \ debugPluginClient.py \ + fenceAgent.py \ guestIF.py \ hooking.py \ hooks.py \ @@ -52,7 +53,8 @@ tc.py \ vdsmDebugPlugin.py \ vmChannels.py \ - vm.py + vm.py \ + $(NULL) dist_vdsmpylib_PYTHON = \ __init__.py \ diff --git a/vdsm/clientIF.py b/vdsm/clientIF.py index 8ba25a7..8760437 100644 --- a/vdsm/clientIF.py +++ b/vdsm/clientIF.py @@ -96,6 +96,7 @@ self.lastRemoteAccess = 0 self._memLock = threading.Lock() self._enabled = True + self.shutdownEvent = threading.Event() self._netConfigDirty = False self._prepareMOM() threading.Thread(target=self._recoverExistingVms, @@ -219,6 +220,7 @@ for binding in self.bindings.values(): binding.prepareForShutdown() self._enabled = False + self.shutdownEvent.set() self.channelListener.stop() self._hostStats.stop() if self.mom: -- To view, visit http://gerrit.ovirt.org/7190 To unsubscribe, visit http://gerrit.ovirt.org/settings Gerrit-MessageType: newchange Gerrit-Change-Id: I944c6548a42612f705a410fb4290215451bca035 Gerrit-PatchSet: 1 Gerrit-Project: vdsm Gerrit-Branch: master Gerrit-Owner: Saggi Mizrahi <smizrahi(a)redhat.com>

7 23

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

vdsm-patches November 2012