Deadlock occures in receiverd
Bug #1580589 reported by
Alexander Kislitsky
This bug affects 1 person
| Affects | Status | Importance | Assigned to | Milestone | |
|---|---|---|---|---|---|
| Fuel for OpenStack |
Confirmed
|
Medium
|
Fuel Sustaining | ||
| Mitaka |
Won't Fix
|
Medium
|
Fuel Sustaining | ||
Bug Description
We have deadlocks introduced into receiverd. For catching the root cause we should find the place there locking order is broken. It is very helpful to set the following values to PostgreSQL settings in postgresql.conf:
log_statement = 'all'
log_line_prefix = '%t %x [%p-%l] '
We should build custom ISO with configured PostgreSQL. We should run repeating
custom system test against the custom ISO util deadlock appers in the logs.
After that we should analyze the DB logs and locking order in the transactions
with the deadlock (grep by the %x param in log_line_prefix).
| Changed in fuel: | |
| importance: | High → Medium |
| status: | New → Confirmed |
Revision history for this message
| OpenStack Infra (hudson-openstack) wrote Change abandoned on fuel-qa (master) | #2 |
Revision history for this message
| OpenStack Infra (hudson-openstack) wrote Change abandoned on fuel-library (master) | #3 |
To post a comment you must log in.
Related changes:
https:/
https:/