Please forgive me if this isn't the right format for this request; it's been a long time since I've done one of these.
I need to push the following change to the review-stats code through puppet:
diff --git a/modules/review-stats/files/review-stats.py b/modules/review-stats/files/review-stats.py index 2c59752..e494168 100755 --- a/modules/review-stats/files/review-stats.py +++ b/modules/review-stats/files/review-stats.py @@ -205,7 +205,7 @@ def run_query(bz): bugdata[bug.id]['hidden'] = 1
# Get the status of each "interesting" bug - for i in seq_max_split(alldeps, 500): + for i in seq_max_split(alldeps, 100): for bug in filter(None, bz._proxy.Bug.get_bugs({'ids':i, 'permissive': 1})['bugs']): interesting[bug['id']] = bug
RHBugzilla is pretty broken right now, and I'm working on fixing this code upstream in a better way which I will push once we unfreeze, but this is the minimal fix that appears to at least get the script to complete without throwing proxy errors. It takes ages to run but it does eventually finish.
- J<
On Fri, 24 May 2013 11:44:53 -0500 "Jason L Tibbitts III" tibbs@math.uh.edu wrote:
Please forgive me if this isn't the right format for this request; it's been a long time since I've done one of these.
I need to push the following change to the review-stats code through puppet:
diff --git a/modules/review-stats/files/review-stats.py b/modules/review-stats/files/review-stats.py index 2c59752..e494168 100755 --- a/modules/review-stats/files/review-stats.py +++ b/modules/review-stats/files/review-stats.py @@ -205,7 +205,7 @@ def run_query(bz): bugdata[bug.id]['hidden'] = 1
# Get the status of each "interesting" bug
- for i in seq_max_split(alldeps, 500):
- for i in seq_max_split(alldeps, 100): for bug in filter(None, bz._proxy.Bug.get_bugs({'ids':i,
'permissive': 1})['bugs']): interesting[bug['id']] = bug
RHBugzilla is pretty broken right now, and I'm working on fixing this code upstream in a better way which I will push once we unfreeze, but this is the minimal fix that appears to at least get the script to complete without throwing proxy errors. It takes ages to run but it does eventually finish.
Seems like a simple change and it's not working at all now, so +1 here.
kevin
On 24 May 2013 10:57, Kevin Fenzi kevin@scrye.com wrote:
On Fri, 24 May 2013 11:44:53 -0500 "Jason L Tibbitts III" tibbs@math.uh.edu wrote:
Please forgive me if this isn't the right format for this request; it's been a long time since I've done one of these.
I need to push the following change to the review-stats code through puppet:
diff --git a/modules/review-stats/files/review-stats.py b/modules/review-stats/files/review-stats.py index 2c59752..e494168 100755 --- a/modules/review-stats/files/review-stats.py +++ b/modules/review-stats/files/review-stats.py @@ -205,7 +205,7 @@ def run_query(bz): bugdata[bug.id]['hidden'] = 1
# Get the status of each "interesting" bug
- for i in seq_max_split(alldeps, 500):
- for i in seq_max_split(alldeps, 100): for bug in filter(None, bz._proxy.Bug.get_bugs({'ids':i,
'permissive': 1})['bugs']): interesting[bug['id']] = bug
RHBugzilla is pretty broken right now, and I'm working on fixing this code upstream in a better way which I will push once we unfreeze, but this is the minimal fix that appears to at least get the script to complete without throwing proxy errors. It takes ages to run but it does eventually finish.
Seems like a simple change and it's not working at all now, so +1 here.
I would say drop it down to 50 for the moment. The 4.4 upgrade is undergoing a lot of changes to try and get it working but it could be a while (and it could get worse before it gets better).
"SJS" == Stephen John Smoogen smooge@gmail.com writes:
SJS> I would say drop it down to 50 for the moment.
I can do that; I just picked 100 because I found that 250 was low enough for things to work so I "halved" it to add some headroom. If setting it to 20% of what works is enough to get me a +1 so I can get this thing running again, great.
- J<
On 24 May 2013 15:00, Jason L Tibbitts III tibbs@math.uh.edu wrote:
"SJS" == Stephen John Smoogen smooge@gmail.com writes:
SJS> I would say drop it down to 50 for the moment.
I can do that; I just picked 100 because I found that 250 was low enough for things to work so I "halved" it to add some headroom. If setting it to 20% of what works is enough to get me a +1 so I can get this thing running again, great.
+1 then
- J<
infrastructure mailing list infrastructure@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/infrastructure
FYI, all this does is submit several smaller lists of tickets to a query. Currently I need to get the open/closed status of 519 tickets; passing a list of 500 to bugzilla used to work but no longer does. Passing 100 seems to work, but still takes ages. Pretty sure I can do it better but I wanted to send a minimal change. This isn't really a critical service anyway.
- J<
infrastructure@lists.fedoraproject.org