Bug 354438 - akonadi-imap-resource extremely slow and 100% CPU usage
Summary: akonadi-imap-resource extremely slow and 100% CPU usage
Status: REPORTED
Alias: None
Product: Akonadi
Classification: Frameworks and Libraries
Component: IMAP resource (show other bugs)
Version: unspecified
Platform: Ubuntu Linux
: NOR grave
Target Milestone: ---
Assignee: Christian Mollekopf
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2015-10-27 07:07 UTC by Vasilis Tsiligiannis
Modified: 2019-06-25 15:54 UTC (History)
5 users (show)

See Also:
Latest Commit:
Version Fixed In:

Attachments
Add an attachment

Note You need to log in before you can comment on or make changes to this bug.
Description Vasilis Tsiligiannis 2015-10-27 07:07:31 UTC 
I have a very large IMAP folder and the akonadi-imap-resource takes forever to read it after the latest upgrade to 15.08.2. Previous version read my emails in a few minutes but current takes hours if it ever completes the process. This renders the whole support for IMAP broken since you cannnot receive or view new e-mails.

Reproducible: Always
Comment 1 Vasilis Tsiligiannis 2015-10-28 20:38:04 UTC 
Additionally, I noticed that akonadi_imap_resource process consume huge amounts of memory (~ 360MB).
Comment 2 Dennis Schridde 2017-04-22 14:16:35 UTC 
I have a similar issue with 17.04.0: `htop` showed multiple threads of `akonadi_imap_resource` running with very high CPU utilisation (50% for several hours apparently). Then suddenly CPU utilisation spiked to 100% for one thread and 200% for the process and memory usage started to climb rapidly until the process was OOM killed after a few minutes at well above 10GiB RAM usage.
Comment 3 Viorel-Cătălin Răpițeanu 2019-01-22 08:49:26 UTC 
I have the same issue on the latest Plasma/Frameworks/Applications.
Comment 4 carl 2019-06-05 18:35:21 UTC 
I still have the 100% cpu problem with KMail 19.04. 


But not always, after killing KMail (I didn't restart Akonadi) and restarting it, I get a normal ~10-13% cpu.

I will try to find a way to reproduce this bug.
Comment 5 Christoph Feck 2019-06-25 15:54:32 UTC 
Do the workarounds for bug 373443 help also for this issue?