[Freeipa-devel] slow response

[Stephen Ingram]

On Tue, Jul 24, 2012 at 11:09 AM, John Dennis <jdennis at redhat.com> wrote:
> On 07/24/2012 12:03 PM, Stephen Ingram wrote:
>>
>> On Mon, Jul 23, 2012 at 11:25 AM, John Dennis <jdennis at redhat.com> wrote:
>>>
>>> On 07/23/2012 12:37 PM, John Dennis wrote:
>>>>
>>>>
>>>> Note the timestamps only have 1 second resolution, something
>>>> which would be easy to fix, so bear that in mind when looking at the
>>>> output of the script.
>>>
>>>
>>>
>>> This is partially as a note to myself, but also to others who want to
>>> test.
>>> The following patch should produce timestamps in the debug output with
>>> microsecond resolution.
>>
>>
>> No one seems to be jumping in on this. Should I run the test again
>> with this patch? I'm really interested in getting this resolved as I
>> can't help but think others are experiencing the same problem. I do
>> have a user in the system that is scheduled for deletion. Would it
>> help for you to grab a ticket with those credentials so you could
>> actually see the slowness in action?
>
>
> It would be good to re-run your test with better timing information.
> Attached is a patch that adds more timing information to the debug log
> output. Please apply the patch and re-run your test the same way you did
> before and email me the contents of /var/log/httpd/error_log. In the
> error_log you should see lines with "elapsed_time=xxxx", e.g.
>
> INFO: admin at XXX.XXX.XXX.XXX.XX: batch(({'params': ((), {}), 'method':
> 'ping'},)): SUCCESS elapsed_time=0.001663
>
> To apply the patch you'll need to be root to modify the files, this should
> work:
>
> % su
> % cd /usr/lib/python2.7/site-packages
> % patch -p1 -b < path_to_saved_patch_file
>
> If you don't have patch installed:
>
> % sudo yum install patch
>
> Then restart httpd and re-run the test.

Seeing python2.7, I'm guessing these patches were against Fedora.
Since I couldn't get them to apply against RH 6.3 I applied them by
hand. I couldn't get the WebUI to load after applying the patches. I'm
not sure of the code that caused the problem, but I did see mention of
global name datetime not being defined. You wouldn't happen to have
patches for RH 6.3?

Steve