rewriter seems to negative cache for a long time

Bug #411250 reported by Steve McInerney on 2009-08-10
This bug affects 1 person
Affects Status Importance Assigned to Milestone
Launchpad itself

Bug Description

This is a bit WTF/guessing atm.

We have a more aggressive check on staging codebrowse these days.
We've noticed that with the daily staging updates; this check goes into an error state due to a 302 redirection (Did I mention it was more aggressive?).

Restarting staging codebrowse, makes no difference to the failure.
graceful'ling apache - and hence restarting the rewrite, seems to fix the problem.

It *feels* like a negative cache issue; but at this stage I'm educated guessing from two separate failures, and triaging from how I got things working again.
Unsure if this is even likely to be an issue in production; suspecting not, but just in case...

Steve McInerney (spm) wrote :

Had same again today. graceful'ling (restarting rewrite) fixed the problem.

Tim Penhey (thumper) wrote :

Steve, is this still an issue now?

Changed in launchpad-code:
status: New → Incomplete
Steve McInerney (spm) wrote :

Haven't noticed it - which is a positive sign.
Last graceful was ~ 27 hours ago.

Fixed or gone away on it's own?

Michael Hudson-Doyle (mwhudson) wrote :

The branch rewrite does not negatively cache, at all. So I don't know what happened here, but it wasn't that :-p

Tom Haddon (mthaddon) on 2010-05-28
tags: added: canonical-losa-lp
Robert Collins (lifeless) wrote :

Does the rewriter log in any fashion whats going on (so that we could start debugging this?)

Changed in launchpad:
status: Incomplete → Triaged
importance: Undecided → High
Robert Collins (lifeless) wrote :

@spm is this still happening?

To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers