Linking back to a particular informational build that took longer without failing:
https://uberchromegw.corp.google.com/i/chromeos.chrome/builders/x86-alex-tot-chrome-pfq-informational/builds/19375

and a particular chrome pfq one that did:
https://uberchromegw.corp.google.com/i/chromeos/builders/x86-alex-chrome-pfq/builds/356/steps/VMTest%20%28attempt%201%29/logs/stdio

The timeout is currently 60 minutes *per test*. The timeout is in fact the same for both builders (and is hard coded, there's no way to differentially change it for different configs), but for some reason on the chromeos.chrome builder the tests came in a little bit under that time (~2800s for the longer test) while I guess on the pfq builder they did not.

Perhaps the pfq builder is doing something else in parallel with vmtest that is slowing it down a little bit. When the vm test time increased, the timeout was marginally being passed on the informational builder, but being failed on the pfq builder.

Not sure what we could have done here, other than making the workload of the builder more similar to the pfq builder.

I'm leaving this open in case we see something similar again, but at the moment I don't think this requires action.

It's been a few weeks, closing, but feel free to reopen if you disagree.