489722 - Steady Ts creep on boxset and cb-minibinus01 since Breakpad

Status: RESOLVED INCOMPLETE

(Camino Graveyard :: General, defect)

Description

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

Following the initial Ts hit, there's been a slow, steady upward creep in Ts on boxset and minus since Breakpad was enabled on 4-15.  (cb-x1, on the other hand, took the initial hit, dropped down a bit, and seems to have remained reasonably constant thereafter.)

Do we have any idea what might be causing this, and if so whether we can make it stop?  

Note I've rebooted minus once since that time, on 4-19 at 14:30 (http://tinderbox.mozilla.org/showbuilds.cgi?tree=Camino&maxdate=1240241208&hours=24&legend=0&norules=1), and that may have stemmed the creep there.  (minus has also been plagued with much-more-frequent random orange since Breakpad; tonight I upped Ts timeout locally to see if that cuts down on the number of oranges.)

Comment 1

[Smokey Ardisson (offline for a while; not following bugs - do not email)]

(In reply to comment #0)
> Following the initial Ts hit, there's been a slow, steady upward creep in Ts on
> boxset and minus since Breakpad was enabled on 4-15.  (cb-x1, on the other
> hand, took the initial hit, dropped down a bit, and seems to have remained
> reasonably constant thereafter.)

Either the backout for bug 475201, or Sam's reboot before the backout for bug 475201, temporarily reduced Ts on boxset to what it was just after Breakpad landed, but it continues to creep upward slowly.  minus has started creeping slowly upward again, too.

> (minus has also been plagued with
> much-more-frequent random orange since Breakpad; tonight I upped Ts timeout
> locally to see if that cuts down on the number of oranges.)

Bumping the Ts timeout on minus to 30s did not help appreciably.  Last night I bumped it again to 50s and we've been green 24hrs.  I'm going to bump it back down to 40s tonight and see if that value is OK or not.