Use SIGKILL by default in Sidekiq::MemoryKiller
This makes the memory growth-triggered Sidekiq restarts more reliable by reducing the chance that Sidekiq ends up in a state where it is not accepting new jobs but also not shutting down: SIGKILL is more likely to work than SIGTERM.
This commit is contained in:
parent
091c906131
commit
c291ff9c6f
|
@ -57,6 +57,7 @@ v 7.11.0 (unreleased)
|
|||
- Show user roles by comments.
|
||||
- Fix automatic blocking of auto-created users from Active Directory.
|
||||
- Call merge request web hook for each new commits (Arthur Gautier)
|
||||
- Use SIGKILL by default in Sidekiq::MemoryKiller
|
||||
|
||||
v 7.10.2
|
||||
- Fix CI links on MR page
|
||||
|
|
|
@ -7,7 +7,7 @@ module Gitlab
|
|||
GRACE_TIME = (ENV['SIDEKIQ_MEMORY_KILLER_GRACE_TIME'] || 15 * 60).to_s.to_i
|
||||
# Wait 30 seconds for running jobs to finish during graceful shutdown
|
||||
SHUTDOWN_WAIT = (ENV['SIDEKIQ_MEMORY_KILLER_SHUTDOWN_WAIT'] || 30).to_s.to_i
|
||||
SHUTDOWN_SIGNAL = (ENV['SIDEKIQ_MEMORY_KILLER_SHUTDOWN_SIGNAL'] || 'SIGTERM').to_s
|
||||
SHUTDOWN_SIGNAL = (ENV['SIDEKIQ_MEMORY_KILLER_SHUTDOWN_SIGNAL'] || 'SIGKILL').to_s
|
||||
|
||||
# Create a mutex used to ensure there will be only one thread waiting to
|
||||
# shut Sidekiq down
|
||||
|
|
Loading…
Reference in New Issue