New issue
Advanced search Search tips

Issue 630366 link

Starred by 2 users

Issue metadata

Status: WontFix
Owner: ----
Closed: Feb 2018
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: ----
Pri: 2
Type: Bug

Blocked on:
issue 600469



Sign in to add a comment

'git cl try' fails with obscure error if latest patchset is deleted

Project Member Reported by kulshin@chromium.org, Jul 21 2016

Issue description

I created a new issue with several patchsets. I then deleted the latest patchset. When I then tried to run 'git cl try' (expecting it to run a tryjob for the latest existing patchset), I got the following error:
===
WARNING: failed to trigger CQ Dry Run.
Either:
 * your project has no CQ
 * you don't have permission to trigger Dry Run
 * bug in this code (see stack trace below).
Consider specifying which bots to trigger manually or asking your project owners for permissions or contacting Chrome Infrastructure team at https://www.chromium.org/infra


Access to https://codereview.chromium.org is denied (server returned HTTP 403).
===

 
Cc: tandrii@chromium.org
Components: -Infra Infra>CQ
Do you have a stack trace?
There was no stack trace provided in the output. This was on Windows, if that matters.
Components: -Infra>CQ Infra>SDK
Status: Available (was: Untriaged)
Thanks for report! It bad that there is no stack trace. I didn't think of this scenario, tbh.

You clearly assumed that of codereview state is primary source of truth, while others consider their checkouts to be primary. That's why triggering tryjobs on "latest" patchset is ambiguous.

Do you think it'd be reasonable to issue a warning instead of guessing, smth like:

 It seems your checkout is not in sync with codereview:
   latest patchset #10001 in checkout, but #8001 in codereview
 Do you want to trigger tryjobs on latest patchset in codereview? Y/N
A warning would be reasonable. That would alert the developer that they may be doing something unintended and help them figure out what they should do.
Labels: -Restrict-View-Google
Labels: Proj-Gerrit-Migration
Labels: Milestone-Dogfood

Comment 8 by aga...@chromium.org, Jan 12 2017

Labels: -Proj-Gerrit-Migration -Milestone-Dogfood
Not sure why this is in the gerrit migration list; it doesn't seem to have anything to do with gerrit. The original report was using rietveld, and gerrit (by design) doesn't offer deletion of patchsets in the UI.

Removing from our work list unless someone can explain why it should be here.
Blockedon: 600469
Agree. In fact, this will be WontFix after  issue 600469  is done.
Status: WontFix (was: Available)

Sign in to add a comment