Re: [ClusterLabs] Antw: Re: Using different folder for /var/lib/pacemaker and usage of /dev/shm files

2016-05-17 Thread Nikhil Utane
Yes, we do have our application using shared memory which is what we see when the cluster is down. On Tue, May 17, 2016 at 10:53 PM, Ken Gaillot wrote: > On 05/17/2016 12:02 PM, Nikhil Utane wrote: > > OK. Will do that. > > > > Actually I gave the /dev/shm usage when the

Re: [ClusterLabs] Antw: Re: Using different folder for /var/lib/pacemaker and usage of /dev/shm files

2016-05-17 Thread Ken Gaillot
On 05/17/2016 12:02 PM, Nikhil Utane wrote: > OK. Will do that. > > Actually I gave the /dev/shm usage when the cluster wasn't up. > When it is up, I see it occupies close to 300 MB (it's also the DC). Hmmm, there should be no usage if the cluster is stopped. Any memory used by the cluster will

Re: [ClusterLabs] Antw: Re: Using different folder for /var/lib/pacemaker and usage of /dev/shm files

2016-05-17 Thread Nikhil Utane
OK. Will do that. Actually I gave the /dev/shm usage when the cluster wasn't up. When it is up, I see it occupies close to 300 MB (it's also the DC). tmpfs 500.0M329.4M170.6M 66% /dev/shm On another node the same is 115 MB. Anyways, I'll monitor the usage to know

Re: [ClusterLabs] Antw: Re: Using different folder for /var/lib/pacemaker and usage of /dev/shm files

2016-05-17 Thread Nikhil Utane
What I would like to understand is how much total shared memory (approximately) would Pacemaker need so that accordingly I can define the partition size. Currently it is 300 MB in our system. I recently ran into insufficient shared memory issue because of improper clean-up. So would like to

[ClusterLabs] Antw: Re: Using different folder for /var/lib/pacemaker and usage of /dev/shm files

2016-05-17 Thread Ulrich Windl
Hi! One of the main problems I identified with POSIX shared memory (/dev/shm) in Linux is that changes to the shared memory don't affect the i-node, so you cannot tell from a "ls -rtl" which segments are still active and which are not. You can only see the creation time. Maybe there should be