-
Notifications
You must be signed in to change notification settings - Fork 4.9k
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
[Heartbeat] Better defaults for browser concurrency #32082
Comments
Pinging @elastic/uptime (Team:Uptime) |
@emilioalvap pointed out that in containers this is tricky to impossible, so, we should probably just default this number to |
Looking at docker on my system I propose that we:
Any objection? Pulled from this SO answer: https://stackoverflow.com/questions/42187085/check-mem-limit-within-a-docker-container |
Not an objection but adding an example as to why I'm wary of this feature.
For reference, this tool implements a similar check: https://github.com/freeipa/freeipa/blob/3237ade3d2df20c3aeba4405f46a45a2130fbc7e/ipaserver/install/installutils.py#L1084 |
Fixes elastic#32082 by limiting browser jobs to 2
Fixes #32082 by limiting browser jobs to 2. Users can still override this with the global heartbeat.jobs.browser.limit: 42. This also fixes a previously unknown bug, where changing the heartbeat.jobs.*.limit values would cause heartbeat to crash with a panic. This appears to be a bug ingo-ucfg. The workaround this PR adds is moving from map[string]JobLimit to map[string]*JobLimit, which doesn't trigger the same reflection issues.
Fixes #32082 by limiting browser jobs to 2. Users can still override this with the global heartbeat.jobs.browser.limit: 42. This also fixes a previously unknown bug, where changing the heartbeat.jobs.*.limit values would cause heartbeat to crash with a panic. This appears to be a bug ingo-ucfg. The workaround this PR adds is moving from map[string]JobLimit to map[string]*JobLimit, which doesn't trigger the same reflection issues. (cherry picked from commit cdd37ca)
Fixes elastic#32082 by limiting browser jobs to 2
Fixes #32082 by limiting browser jobs to 2. Users can still override this with the global heartbeat.jobs.browser.limit: 42. This also fixes a previously unknown bug, where changing the heartbeat.jobs.*.limit values would cause heartbeat to crash with a panic. This appears to be a bug ingo-ucfg. The workaround this PR adds is moving from map[string]JobLimit to map[string]*JobLimit, which doesn't trigger the same reflection issues. (cherry picked from commit cdd37ca) Co-authored-by: Andrew Cholakian <[email protected]>
Fixes #32082 by limiting browser jobs to 2. Users can still override this with the global heartbeat.jobs.browser.limit: 42. This also fixes a previously unknown bug, where changing the heartbeat.jobs.*.limit values would cause heartbeat to crash with a panic. This appears to be a bug ingo-ucfg. The workaround this PR adds is moving from map[string]JobLimit to map[string]*JobLimit, which doesn't trigger the same reflection issues.
When running browser monitors we should automatically limit the number of concurrent browser monitors based on available system memory, requiring 750MiB free per monitor.
This should limit problems when running monitors via fleet, where concurrently running monitors can quickly trigger an OOM.
See also #23687 .
The text was updated successfully, but these errors were encountered: