The site works again: after a release of a new lab-app version the on-call engineer brought back the previous one. Reviewing the outage takes a timeline, and nobody kept one as it went. Rebuild it from the machine's journal.
Put together the outage timeline
Just now the site answered 502 for a few seconds: after a release of a new lab-app version the on-call engineer brought back the previous one. Nobody kept a timeline. Look at the machine's journal for 15 minutes and write to ~/timeline.txt, as “time event” lines, when the site stopped answering and when it worked again.
Goals · 0 of 1
- Write the start and the end of the outage to ~/timeline.txt not done yet
When did the site begin to answer 502, according to the journal?
-
That is the end of the outage: right after the on-call engineer's commands the journal has RESOLVED.
-
After the copy the old process went on serving requests for six more seconds: FIRING comes after the restart, not after
cp.
The site began to answer 502 right after the lab-app restart; after the file was copied it still worked.