Hi Giacomo,
sorry for a late reply.
It turned out that the major improvement was made through increasing
NFS packet size from 8k to 32k.
After that no more of those messages were shown and everything is
working just great :-)
Thank you everyone for your suggestions and ideas again.
Matej
201
On 09/01/15 13:01, Matej Mailing wrote:
we have tweaked some NFS parameters for our backend storage, including
packet sizes and now the load is constantly slower, there is much less
I/O waiting. There were no lock-ups after that. So it really seems it
was slow storage that was to be blamed. I wil
2015-01-07 13:07 GMT+01:00 Ritesh Raj Sarraf :
> On 01/07/2015 04:32 PM, Tim Bell wrote:
>
>> The profile of this hardware looks virtualized. That makes me ask if you
>> have
>> installed the para-virtualized drivers ?
>>
>
> Out of interest, does using para-virtualized drivers reduce the chance of
On 01/07/2015 04:32 PM, Tim Bell wrote:
>> > The profile of this hardware looks virtualized. That makes me ask if you
>> > have
>> > installed the para-virtualized drivers ?
>> >
> Out of interest, does using para-virtualized drivers reduce the chance of
> soft lockups ? I can understand that it
2015-01-07 9:06 GMT+01:00 Ritesh Raj Sarraf :
>
>
> On 01/05/2015 03:46 PM, Matej Mailing wrote:
>> Hello Erhan,
>>
>> soft lock-up has just happened again on the instance node. I am
>> monitoring network interface traffic on which the NFS server is
>> connected and the interface has constantly bee
> -Original Message-
> From: Ritesh Raj Sarraf [mailto:r...@researchut.com]
> Sent: 07 January 2015 09:06
> To: openstack@lists.openstack.org
> Subject: Re: [Openstack] BUG: soft lockup messages
>
>
> The profile of this hardware looks virtualized. That
2015-01-07 9:06 GMT+01:00 Ritesh Raj Sarraf :
>
>
> On 01/05/2015 03:46 PM, Matej Mailing wrote:
>> Hello Erhan,
>>
>> soft lock-up has just happened again on the instance node. I am
>> monitoring network interface traffic on which the NFS server is
>> connected and the interface has constantly bee
On 01/05/2015 03:46 PM, Matej Mailing wrote:
> Hello Erhan,
>
> soft lock-up has just happened again on the instance node. I am
> monitoring network interface traffic on which the NFS server is
> connected and the interface has constantly been under 20% of it's
> capacity, also the load on the N
2015-01-04 21:20 GMT+01:00 Erhan Ekici :
> Hi Matej,
>
> Can you change the values(both on compute and instance node) and see if soft
> lock-ups generated again? There are two approaches; one decreasing and the
> other increasing the cache.
>
> vm.dirty_background_ratio = 5
> vm.dirty_ratio = 10
>
2015-01-04 21:20 GMT+01:00 Erhan Ekici :
> Hi Matej,
>
> Can you change the values(both on compute and instance node) and see if soft
> lock-ups generated again? There are two approaches; one decreasing and the
> other increasing the cache.
>
> vm.dirty_background_ratio = 5
> vm.dirty_ratio = 10
>
Hi Matej,
Can you change the values(both on compute and instance node) and see if
soft lock-ups generated again? There are two approaches; one decreasing and
the other increasing the cache.
vm.dirty_background_ratio = 5
vm.dirty_ratio = 10
If I/O & Cache is not the source of the problem, we shou
2015-01-04 17:10 GMT+01:00 Erhan Ekici :
> Hi Matej,
>
> I am not sure but most of the soft lock-up messages generated due to I/O
> issues. From your log entries, I see mysqld issues as well as others:
>
> Jan 3 10:41:44 postar kernel: [303154.352793] NMI watchdog: BUG: soft
> lockup - CPU#3 stuc
Hi Matej,
I am not sure but most of the soft lock-up messages generated due to I/O
issues. From your log entries, I see mysqld issues as well as others:
Jan 3 10:41:44 postar kernel: [303154.352793] NMI watchdog: BUG: soft lockup
- CPU#3 stuck for 45s! [mysqld:4601]
Jan 3 10:44:00 postar
2015-01-03 17:56 GMT+01:00 John Griffith :
> On Sat, Jan 3, 2015 at 5:05 AM, Matej Mailing wrote:
>> Hi,
>>
>> we are experiencing randomly-timed soft lockup messages in different
>> instances (running CentOS, Ubuntu, etc.) with different processes and
>> on different compute nodes. I suspect that
On Sat, Jan 3, 2015 at 5:05 AM, Matej Mailing wrote:
> Hi,
>
> we are experiencing randomly-timed soft lockup messages in different
> instances (running CentOS, Ubuntu, etc.) with different processes and
> on different compute nodes. I suspect that access to the cinder
> storage via NFS could be t
Hi,
we are experiencing randomly-timed soft lockup messages in different
instances (running CentOS, Ubuntu, etc.) with different processes and
on different compute nodes. I suspect that access to the cinder
storage via NFS could be the cause of issue, but then perhaps all the
instances would trigg
16 matches
Mail list logo
