Closed Bug 975001 Opened 11 years ago Closed 11 years ago

Log large-difference HAWK clock mismatches

Categories

(Cloud Services Graveyard :: Server: Sync, defect)

x86
macOS
defect
Not set
normal

Tracking

(Not tracked)

VERIFIED FIXED

People

(Reporter: telliott, Unassigned)

Details

(Whiteboard: [qa+])

We should start logging large (>2 min?) time differentials between clients and servers to get an idea of the scope of the mismatch problem. A large change in the frequency of these may indicate a problem on the server side.
Moving to Server:Sync since tokenserver is the one thing in this mix that doesn't speak hawk. Is this covered by the changes in Bug 975214?
Component: Server: Token → Server: Sync
Flags: needinfo?(telliott)
Whiteboard: [qa?]
I think it's supplemental. We should still be tracking the big mismatches, both so we would have an idea of scope if we shortened the hawk window and also because a rise in them might indicate a clock problem on the server.
Flags: needinfo?(telliott)
ah, but you do have the logging in there, so yes, we're covered.
Marking resolved/fixed since this is covered by Bug 975214. This code is now live on stage, we should verify that large timestamp differences do in fact show up in the logs before marking it as verified.
Status: NEW → RESOLVED
Closed: 11 years ago
Resolution: --- → FIXED
Whiteboard: [qa?] → [qa+]
We can't really do this accurately since the host times and log times are different. What is the status of that issue?
ISTM we don't have to have the times lining up, we just need to confirm whether the particular log lines produced by Bug 975214 are actually appearing in the log. (Correlated times would be *better* of course, but not necessary)
I think we are good here.
Status: RESOLVED → VERIFIED
Product: Cloud Services → Cloud Services Graveyard
You need to log in before you can comment on or make changes to this bug.