[pve-devel] clocks not syncronized in ceph

[Alexandre DERUMIER]

mmm, maybe some bios option ?

Do you have your cpus set at "maximum performance" (I have this kind of option in dell servers).

Do you have disabled options of power saving? (like C1E, or dynamic cpu freq, core disabling on inactivity,...).




----- Mail original ----- 

De: "Stefan Priebe - Profihost AG" <s.priebe at profihost.ag> 
À: "Alexandre DERUMIER" <aderumier at odiso.com> 
Cc: pve-devel at pve.proxmox.com 
Envoyé: Jeudi 8 Novembre 2012 09:54:54 
Objet: Re: [pve-devel] clocks not syncronized in ceph 

Am 08.11.2012 09:53, schrieb Alexandre DERUMIER: 
> Does it happen when you have a high load average ? 

No machines are Idle but i can't understand why the clock changes are so 
extreme... 

Stefan 

> ----- Mail original ----- 
> 
> De: "Stefan Priebe - Profihost AG" <s.priebe at profihost.ag> 
> À: "Alexandre DERUMIER" <aderumier at odiso.com> 
> Cc: pve-devel at pve.proxmox.com 
> Envoyé: Jeudi 8 Novembre 2012 09:26:52 
> Objet: Re: [pve-devel] clocks not syncronized in ceph 
> 
> Openntpd output looks like this: 
> 
> Nov 6 23:28:15 cloud1-ceph1 ntpd[1572]: adjusting local clock by -0.325597s 
> Nov 7 00:09:48 cloud1-ceph1 ntpd[1572]: skew change 36.903 exceeds limit 
> Nov 7 00:37:12 cloud1-ceph1 ntpd[1572]: skew change -21.291 exceeds limit 
> Nov 7 01:06:15 cloud1-ceph1 ntpd[1572]: skew change -17.565 exceeds limit 
> Nov 7 01:28:06 cloud1-ceph1 ntpd[1572]: skew change -7.156 exceeds limit 
> 
> Stefan 
> Am 08.11.2012 08:53, schrieb Alexandre DERUMIER: 
>> damn, this is strange... 
>> 
>> are you sure that your ntp daemon is running ? (If I remember good, ntpdate can't run if ntpd is already running) 
>> 
>> # ntpdate 0.debian.pool.ntp.org 
>> 8 Nov 08:53:16 ntpdate[731962]: the NTP socket is in use, exiting 
>> 
>> ----- Mail original ----- 
>> 
>> De: "Stefan Priebe - Profihost AG" <s.priebe at profihost.ag> 
>> À: pve-devel at pve.proxmox.com 
>> Envoyé: Jeudi 8 Novembre 2012 08:38:49 
>> Objet: [pve-devel] clocks not syncronized in ceph 
>> 
>> Hello, 
>> 
>> on my ceph cluster i get always the message clocks not syncronized. But 
>> i'm running openntpd on all nodes. 
>> 
>> Time is always different like this: 
>> 0.050004s in the future, clocks not synchronized 
>> 
>> I can also run ntpdate manual multiple times and it will syncronize 
>> again and again. 
>> 
>> /root/sshall "ntpdate pool.ntp.org" 
>> 8 Nov 08:37:47 ntpdate[14094]: adjust time server 212.83.33.85 offset 
>> 0.003316 sec 
>> 8 Nov 08:37:55 ntpdate[12228]: adjust time server 212.83.33.85 offset 
>> 0.005892 sec 
>> 8 Nov 08:38:04 ntpdate[14418]: adjust time server 131.188.3.221 offset 
>> -0.008360 sec 
>> .. 
>> 
>> Greets, 
>> Stefan 
>> _______________________________________________ 
>> pve-devel mailing list 
>> pve-devel at pve.proxmox.com 
>> http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel 
>>