-
Notifications
You must be signed in to change notification settings - Fork 138
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
decrement server_job->function->job_total only if it > 0 #302
Conversation
I was considering this same change, but I was worried it didn’t address the underlying problem. I think it’s worth a try though.
|
This doesn't seem to have completely solved the problem we mentioned in #301 (comment) however like we said before, it's possible this a completely different thing from our original issue: We have had it take a few goes to crash out with the worker than doesn't timeout for some reason. |
…ly if it > 0 this aims to prevent the value went out of the uint32_t range See gearman#301 (comment)
…ly if it > 0 this aims to prevent the value went out of the uint32_t range See gearman#301 (comment)
this aims to prevent the value went out of the uint32_t range See gearman#301 (comment)
… only if it > 0 this aims to prevent the value went out of the uint32_t range See gearman#301 (comment)
if it > 0 this aims to prevent the value went out of the uint32_t range See gearman#301 (comment)
Well, you could add similar guards wherever job_running is decremented. There are only two locations: gearmand/libgearman-server/job.cc Line 301 in e2d76cf
gearmand/libgearman-server/job.cc Line 385 in e2d76cf
But I think gearmand will still hang. Could you try that, @infraweavers ? In order to get to -1, job_running had to have been decremented more than once, obviously. Is gearmand/libgearman-server/job.cc Line 385 in e2d76cf
entirely. |
I did it yesterday, but the result wasn't much better:-(
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This may do more harm than good. I'd like a regression test before we act on the bug for every user.
THis PR does at least seem to stop the |
Going to close this, but feel free to ping / re-open if you can add a regression test. I am still thinking there's something deeper and this is just treating the symptom. |
Yeah sounds good |
The PR aims to fix #301. My tests are based on #301 (comment)