Re: [ceph-users] Crushmap (tunables) flapping on cluster

2017-01-10 Thread Breunig, Steve (KASRL)
Yes, that was the problem, thx.


Von: Stillwell, Bryan J 
Gesendet: Dienstag, 10. Januar 2017 18:06:10
An: Breunig, Steve (KASRL); ceph-users@lists.ceph.com
Betreff: Re: [ceph-users] Crushmap (tunables) flapping on cluster

On 1/10/17, 2:56 AM, "ceph-users on behalf of Breunig, Steve (KASRL)"
 wrote:

>Hi list,
>
>
>I'm running a cluster which is currently in migration from hammer to
>jewel.
>
>
>Actually i have the problem, that the tunables are flapping and a map of
>an rbd image is not working.
>
>
>It is flapping between:
>
>
>{
>"choose_local_tries": 0,
>"choose_local_fallback_tries": 0,
>"choose_total_tries": 50,
>"chooseleaf_descend_once": 1,
>"chooseleaf_vary_r": 1,
>"chooseleaf_stable": 0,
>"straw_calc_version": 1,
>"allowed_bucket_algs": 54,
>"profile": "hammer",
>"optimal_tunables": 0,
>"legacy_tunables": 0,
>"minimum_required_version": "hammer",
>"require_feature_tunables": 1,
>"require_feature_tunables2": 1,
>"has_v2_rules": 0,
>"require_feature_tunables3": 1,
>"has_v3_rules": 0,
>"has_v4_buckets": 1,
>"require_feature_tunables5": 0,
>"has_v5_rules": 0
>}
>
>
>and
>
>
>{
>"choose_local_tries": 0,
>"choose_local_fallback_tries": 0,
>"choose_total_tries": 50,
>"chooseleaf_descend_once": 1,
>"chooseleaf_vary_r": 1,
>"straw_calc_version": 1,
>"allowed_bucket_algs": 54,
>"profile": "hammer",
>"optimal_tunables": 0,
>"legacy_tunables": 0,
>"require_feature_tunables": 1,
>"require_feature_tunables2": 1,
>"require_feature_tunables3": 1,
>"has_v2_rules": 0,
>"has_v3_rules": 0,
>"has_v4_buckets": 1
>}
>
>
>Did someone have that problem too?
>How can it be solved?

Have you upgraded all the mon nodes?  My guess is that when you're running
'ceph osd crush show-tunables' it's sometimes being reported from a hammer
mon node and sometimes from a jewel mon node.

You can run 'ceph tell mon.* version' to verify they're all running the
same version.

When you say the map is failing, are you using the kernel rbd driver?  If
so you might need to upgrade your kernel to support the new features in
jewel.

Bryan

E-MAIL CONFIDENTIALITY NOTICE:
The contents of this e-mail message and any attachments are intended solely for 
the addressee(s) and may contain confidential and/or legally privileged 
information. If you are not the intended recipient of this message or if this 
message has been addressed to you in error, please immediately alert the sender 
by reply e-mail and then delete this message and any attachments. If you are 
not the intended recipient, you are notified that any use, dissemination, 
distribution, copying, or storage of this message or any attachment is strictly 
prohibited.



Kantar Disclaimer<http://www.kantar.com/disclaimer.html>
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com


[ceph-users] Crushmap (tunables) flapping on cluster

2017-01-10 Thread Breunig, Steve (KASRL)
Hi list,


I'm running a cluster which is currently in migration from hammer to jewel.


Actually i have the problem, that the tunables are flapping and a map of an rbd 
image is not working.


It is flapping between:

{
"choose_local_tries": 0,
"choose_local_fallback_tries": 0,
"choose_total_tries": 50,
"chooseleaf_descend_once": 1,
"chooseleaf_vary_r": 1,
"chooseleaf_stable": 0,
"straw_calc_version": 1,
"allowed_bucket_algs": 54,
"profile": "hammer",
"optimal_tunables": 0,
"legacy_tunables": 0,
"minimum_required_version": "hammer",
"require_feature_tunables": 1,
"require_feature_tunables2": 1,
"has_v2_rules": 0,
"require_feature_tunables3": 1,
"has_v3_rules": 0,
"has_v4_buckets": 1,
"require_feature_tunables5": 0,
"has_v5_rules": 0
}

and

{
"choose_local_tries": 0,
"choose_local_fallback_tries": 0,
"choose_total_tries": 50,
"chooseleaf_descend_once": 1,
"chooseleaf_vary_r": 1,
"straw_calc_version": 1,
"allowed_bucket_algs": 54,
"profile": "hammer",
"optimal_tunables": 0,
"legacy_tunables": 0,
"require_feature_tunables": 1,
"require_feature_tunables2": 1,
"require_feature_tunables3": 1,
"has_v2_rules": 0,
"has_v3_rules": 0,
"has_v4_buckets": 1
}


Did someone have that problem too?

How can it be solved?


Regards,

Steve



Kantar Disclaimer
___
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com