|
|
Created:
5 years, 10 months ago by dvadym Modified:
5 years, 10 months ago CC:
chromium-reviews, gcasto+watchlist_chromium.org, mkwst+watchlist-passwords_chromium.org Base URL:
https://chromium.googlesource.com/chromium/src.git@master Target Ref:
refs/pending/heads/master Project:
chromium Visibility:
Public. |
DescriptionPostpone collecting passwords UMA metrics.
BUG=453835
Committed: https://crrev.com/f2d4cd89181eea9a913487b5fe5c9f0a432955cf
Cr-Commit-Position: refs/heads/master@{#314350}
Patch Set 1 #Patch Set 2 : Fix #
Total comments: 2
Patch Set 3 : Small enhancement #
Total comments: 1
Messages
Total messages: 18 (5 generated)
dvadym@chromium.org changed reviewers: + vabr@chromium.org
Hi Vaclav, Could you please review this CL? Best regards, Vadym
LGTM, this sounds reasonable. I'm still a bit worried about this practice of randomly postponing tasks in general, instead of having recognised events of "load is low, let's do less important stuff". There was a promising proposal about runlevels on chromium-dev in 2013 https://groups.google.com/a/chromium.org/d/msg/chromium-dev/i7nyQOWRUfY/qdJOZ..., but that somehow sadly died out. Cheers, Vaclav
The CQ bit was checked by dvadym@chromium.org
CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/887043004/1
The CQ bit was unchecked by commit-bot@chromium.org
Try jobs failed on following builders: ios_dbg_simulator on tryserver.chromium.mac (http://build.chromium.org/p/tryserver.chromium.mac/builders/ios_dbg_simulator...)
Hi Vaclav, Could you please take another look at this CL? Best regards, Vadym
Thanks for the fix. I should have noted that delaying all PasswordStore tasks by 30 seconds is not the best idea. :) LGTM. Cheers, Vaclav https://codereview.chromium.org/887043004/diff/20001/components/password_mana... File components/password_manager/core/browser/password_store.cc (right): https://codereview.chromium.org/887043004/diff/20001/components/password_mana... components/password_manager/core/browser/password_store.cc:165: if (task_runner.get()) { nit: You can drop ".get()".
On 2015/02/03 15:43:54, vabr (Chromium) wrote: > Thanks for the fix. I should have noted that delaying all PasswordStore tasks by > 30 seconds is not the best idea. :) s/noted/noticed (In other words, sorry for a crappy review.)
Thanks, Vaclav. Sorry, it's also my fault, I should have noticed this also. https://codereview.chromium.org/887043004/diff/20001/components/password_mana... File components/password_manager/core/browser/password_store.cc (right): https://codereview.chromium.org/887043004/diff/20001/components/password_mana... components/password_manager/core/browser/password_store.cc:165: if (task_runner.get()) { On 2015/02/03 15:43:54, vabr (Chromium) wrote: > nit: You can drop ".get()". Done.
The CQ bit was checked by dvadym@chromium.org
CQ is trying da patch. Follow status at https://chromium-cq-status.appspot.com/patch-status/887043004/40001
Message was sent while issue was closed.
Committed patchset #3 (id:40001)
Message was sent while issue was closed.
Patchset 3 (id:??) landed as https://crrev.com/f2d4cd89181eea9a913487b5fe5c9f0a432955cf Cr-Commit-Position: refs/heads/master@{#314350}
Message was sent while issue was closed.
vasilii@chromium.org changed reviewers: + vasilii@chromium.org
Message was sent while issue was closed.
https://codereview.chromium.org/887043004/diff/40001/components/password_mana... File components/password_manager/core/browser/password_store.cc (right): https://codereview.chromium.org/887043004/diff/40001/components/password_mana... components/password_manager/core/browser/password_store.cc:167: base::Bind(&PasswordStore::ReportMetricsImpl, this, sync_username, My concern is that you capture the scoped pointer for 30 seconds. After this 30 seconds the password store may have its Shutdown() called. While it probably doesn't lead to crash, I don't think that delaying PasswordStore and LoginDatabase destruction is a good idea.
Message was sent while issue was closed.
I checked, even in case when we delete user profile PasswordStore is not destructed. So it will not cause any crashes. But there is another strange thing, that even if there are two profiles in use, we still have only report collection. I'll investigate it later. |