There's probably a bottleneck in your config. Either at the hypervisor level or inside the VM.

On my instance, I have 150 tickets/day, up to ~40 agents and up to ~140 customers simultaneously connected. I'm running Apache 32-bit/MySQL 32-bit/Strawberry Perl 32-bit/Windows 2008 R2 64-bit in a VM with 4 CPUs on Hyper-V (Windows 2008 R2), on 4 year old hardware. I think my access times are slightly better than yours.

Ensure your VM performs optimally. It may be worth to install OTRS in a different VM, on a different hypervisor and see if you get the same performance.

Make sure you follow chapter 6 from the admin manual for performance improvements but really for 2-3 customers and 2-3 agents you shouldn't need to do make any special performance performance tweaks for OTRS to runs decently fast.



On Tue, May 13, 2014 at 12:04 PM, Mimiko <vbvbrj@gmail.com> wrote:
On 13.05.2014 11:56, Bogdan Iosif wrote:
Hi Mimiko. Did you manage to solve the performance issue?

Hello.
Somehow. I did what mhillman sugested - ran
shell> bin/otrs.RebuildConfig.pl
shell> bin/otrs.DeleteCache.pl

but in long terms, still it is visible slow for agents. Dashboard generation on server for agents takes 0-6 seconds. Accessing SysConfig or Package Manager takes 5-15 seconds.

In case of customers and agents it seems that if the ticket system is not accessed for some time, then first to access will wait awhile. Then others will access pages quicker. Also, when selecting a queue or service in new ticket page, refreshing other fields takes 1-2 seconds. And this is a starting project, where there are 2-3 customers and 2-3 agents. I think that perl is slow anyway.


--
Mimiko desu.
---------------------------------------------------------------------
OTRS mailing list: otrs - Webpage: http://otrs.org/
Archive: http://lists.otrs.org/pipermail/otrs
To unsubscribe: http://lists.otrs.org/cgi-bin/listinfo/otrs