if it was a different service. Anyway, I would probably start to
compare memcached configs.
Zitat von "Braden, Albert" <
abraden@verisign.com>:
> The timing result is 1.6 seconds (vs 0.2 on a non-rebuilt region),
> but the elapsed time is 2.7 seconds (vs 1.2):
>
> Rebuilt region:
>
> $ time openstack user show --timing user
> +---------------------+------------------------------------------------------------------+
> | Field | Value
> |
> +---------------------+------------------------------------------------------------------+
> | domain_id | <DID> |
> | email |
user@ourdomain.com > |
> | enabled | True
> |
> | id | <ID> |
> | name | user
> |
> | options | {}
> |
> | password_expires_at | None
> |
> +---------------------+------------------------------------------------------------------+
>
> +------------------------------------------------------------------------------------------------------------------+--------------------+
> | URL
> | Seconds |
> +------------------------------------------------------------------------------------------------------------------+--------------------+
> | GET
https://api-ext.region.ourdomain.com:5000/v3 > |
> 0.030608 |
> | POST
https://api-ext.region.ourdomain.com:5000/v3/auth/tokens > |
> 0.550521 |
> | GET
https://api-ext.region.ourdomain.com:5000/ > |
> 0.006306 |
> | GET
https://api-ext.region.ourdomain.com:5000/v3/auth/tokens > |
> 0.608377 |
> | GET
https://api-ext.region.ourdomain.com:5000/v3/users/<ID> |
> 0.458892 |
> | Total
> | 1.6547040000000002 |
> +------------------------------------------------------------------------------------------------------------------+--------------------+
>
> real 0m2.733s
> user 0m0.982s
> sys 0m0.105s
>
> Not rebuilt:
>
> $ time openstack user show --timing user
> +---------------------+------------------------------------------------------------------+
> | Field | Value
> |
> +---------------------+------------------------------------------------------------------+
> | domain_id | <DID> |
> | email |
user@ourdomain.com > |
> | enabled | True
> |
> | id | <ID> |
> | name | user
> |
> | options | {}
> |
> | password_expires_at | None
> |
> +---------------------+------------------------------------------------------------------+
>
> +------------------------------------------------------------------------------------------------------------------+----------+
> | URL
> | Seconds |
> +------------------------------------------------------------------------------------------------------------------+----------+
> | GET
https://api-ext.oldregion.ourdomain.com:5000/v3 > | 0.026787 |
> | POST
https://api-ext.oldregion.ourdomain.com:5000/v3/auth/tokens > | 0.042511 |
> | GET
https://api-ext.oldregion.ourdomain.com:5000/ > | 0.00767 |
> | GET
https://api-ext.oldregion.ourdomain.com:5000/v3/auth/tokens > | 0.049546 |
> | GET
https://api-ext.oldregion.ourdomain.com:5000/v3/users/<ID> | 0.044296 |
> | Total
> | 0.17081 |
> +------------------------------------------------------------------------------------------------------------------+----------+
>
> real 0m1.176s
> user 0m0.892s
> sys 0m0.118s
>
> -----Original Message-----
> From: Eugen Block <
eblock@nde.ag>
> Sent: Tuesday, April 2, 2024 12:52 PM
> To:
openstack-discuss@lists.openstack.org> Subject: [EXTERNAL] Re: [kolla] [Wallaby] Openstack operations 2x
> slower after rebuilding from RHEL8/Train to RHEL9/Wallaby
>
> Caution: This email originated from outside the organization. Do not
> click links or open attachments unless you recognize the sender and
> know the content is safe.
>
> Hi, could you please share the details of (or any other command that
> takes longer than usual):
>
> openstack user show --timing
>
> Thanks,
> Eugen
>
> Zitat von Albert Braden <
ozzzo@yahoo.com>:
>
>> After rebuilding a couple of our clusters, moving from Train on
>> RHEL8 to Wallaby on RHEL9, we see all openstack operations taking 2x
>> longer. Even simple database reads such as "openstack user show"
>> average 1.5 seconds on old regions and 3 seconds on rebuilt regions.
>> I enabled debug and poked around in the logs but nothing is jumping
>> out at me. I was thinking maybe mariadb needed to be tuned differently
>> from RHEL8->9 but if I log into the mariadb container and query the
>> database it seems equally fast; I can do 100 simple queries in a
>> second. Where should I be looking for the source of this delay?