Confirmed users
360
edits
User:Andrew Sutherland/MailNews/GlobalDatabase/ReviewBelievedResolved: Difference between revisions
User:Andrew Sutherland/MailNews/GlobalDatabase/ReviewBelievedResolved (view source)
Revision as of 10:12, 27 January 2009
, 27 January 2009→dmose
(→dmose) |
|||
| Line 374: | Line 374: | ||
=== dmose === | === dmose === | ||
==== startup | ==== startup ==== | ||
* At startup, I'm seeing some errors, even though the two files | * At startup, I'm seeing some errors, even though the two files | ||
| Line 644: | Line 644: | ||
[https://hg.mozilla.org/users/bugmail_asutherland.org/gloda/rev/0250d7e58b7b gone] | [https://hg.mozilla.org/users/bugmail_asutherland.org/gloda/rev/0250d7e58b7b gone] | ||
==== memory issues ==== | |||
<b>(dmose)</b> <i>with the new indexer, I've seen RSIZE get extremely large (> 1G) once. | |||
Perhaps re-indexing my store would trigger that behavior again.</i> | |||
<b>(asuth)</b> by new indexer, do you mean the sweep-based, event-augmented indexer completed in [https://hg.mozilla.org/users/bugmail_asutherland.org/gloda/file/13592e09f549 changeset 13592e09f549] and pushed on the morning of Sunday, Sep 21? If not, then I think we're okay. If so, then it's cause for concern, although I think it would mainly be the garbage collector's fault for being lazy and maybe we can kick it to make it collect more frequently. | |||
<b>(dmose)</b> I did mean the indexer completed in the changeset you referenced, so I suspect we still have some sort of issue. | |||
<b>(asuth status 10/7)</b> I still need to look into this, especially on OS X. | |||
<b>(asuth)</b>Gloda had a problem where it would not close the databases when it was done with them. This would lead to memory issues. That gloda issue has been addressed. | |||