Horde webmail slow performance on migrated hosts

[dss9360]

---------------------------------------------------------------
PRODUCT, VERSION, OPERATING SYSTEM, ARCHITECTURE
Parallels Plesk Panel v10.4.4_build1013111102.18 os_CentOS 5
GenuineIntel, Intel(R) Xeon(R)CPU E5620 @ 2.40GHz
Linux 2.6.18-028stab095.1

PROBLEM DESCRIPTION
Horde webmail accessed on domains that were migrated using the Plesk Migration manager suffers high performance loss. Loss can be defined as 15-20 seconds to perform such tasks as logging on, navigating menus in webmail, and starting to compose a message.

STEPS TO REPRODUCE
Migrate domain using Plesk Migration Manager.
Old host information: Parallels Pro Control Panel , 10.3.2-rhel.5ES.6 i386

ACTUAL RESULT
Once domain is active, access webmail and performance is hindered greatly.

EXPECTED RESULT
Webmail performance should be reasonable.

ANY ADDITIONAL INFORMATION
It is worth noting that old Parallels host was using squirrelmail as web client.
--------------------------------------------------------------

[dss9360]

If I update the password for an email account (used same password) then horde speeds up to normal. I don't want to do this individually, however.

[LinqLOL]

I can confirm this behavior! We just did one Ensim/PPCP migration (from 10.3.4 to Plesk 11.09) and a lot of people are complaining about slow Horde. When they change their password the problem is solved. I will try to check if it's Horde OR imap related..

Beside of manual password changing is there a work-around?

[IgorG]

Any related errors from Horde and Plesk logs?

[abdi]

What also is your #sestatus?

[Tozz]

I can confirm I have the same problem. After migrating from a custom installation to Plesk (migration is done using the Plesk CLI tools), e-mail accounts from this single domain are very slow when using Horde. Other domains work fine. Setting the password in Plesk webui again resolves the issue. I have been unable to find a cause for this.

I tried /opt/psa/admin/bin/mchk --without-spam and /opt/psa/admin/bin/httpdmng --reconfigure-all. Both had no effect.

Horde logs and Plesk Panel logs do not show any problems.