Hello -
In Prism on an 8-node cluster, we are getting an alert under Health showing 3 of 8 CVMs that their Memory Usage are @ 100%, for about a week now as shown in the graph (in Prism). However, those CVMs seem fine in vCenter. At a high level, Guest Mem % for the CVMs are around 40% or less.
Thoughts?
Thanks.
Page 1 / 1
Hi rvillarin,Could you please get the output of the below command from one of the CVMs showing high memory usage:# cat /proc/meminfoWe will check if this is a cause for worry or not.
What version is your cluster running? I had a similar problem where the health check would fire off into yellow status about high CVM memory usage and get "stuck" - the CVM memory usage would go back down and stay down but the alert would remain triggered for possibly days.
The cluster is running 4.0.1.1.
It's still showing the alert in Prism for weeks now.
Here's the output of cat /proc/meminfo:
MemTotal: 16466964 kBMemFree: 937664 kBBuffers: 272600 kBCached: 1109760 kBSwapCached: 0 kBActive: 14200700 kBInactive: 863396 kBActive(anon): 13682816 kBInactive(anon): 2180 kBActive(file): 517884 kBInactive(file): 861216 kBUnevictable: 24536 kBMlocked: 8184 kBSwapTotal: 0 kBSwapFree: 0 kBDirty: 812 kBWriteback: 0 kBAnonPages: 13706708 kBMapped: 436092 kBShmem: 352 kBSlab: 206088 kBSReclaimable: 108460 kBSUnreclaim: 97628 kBKernelStack: 6768 kBPageTables: 89756 kBNFS_Unstable: 0 kBBounce: 0 kBWritebackTmp: 0 kBCommitLimit: 8233480 kBCommitted_AS: 15552704 kBVmallocTotal: 34359738367 kBVmallocUsed: 302668 kBVmallocChunk: 34359372536 kBHardwareCorrupted: 0 kBAnonHugePages: 11448320 kBHugePages_Total: 0HugePages_Free: 0HugePages_Rsvd: 0HugePages_Surp: 0Hugepagesize: 2048 kBDirectMap4k: 10240 kBDirectMap2M: 16766976 kB
Thanks.
It's still showing the alert in Prism for weeks now.
Here's the output of cat /proc/meminfo:
MemTotal: 16466964 kBMemFree: 937664 kBBuffers: 272600 kBCached: 1109760 kBSwapCached: 0 kBActive: 14200700 kBInactive: 863396 kBActive(anon): 13682816 kBInactive(anon): 2180 kBActive(file): 517884 kBInactive(file): 861216 kBUnevictable: 24536 kBMlocked: 8184 kBSwapTotal: 0 kBSwapFree: 0 kBDirty: 812 kBWriteback: 0 kBAnonPages: 13706708 kBMapped: 436092 kBShmem: 352 kBSlab: 206088 kBSReclaimable: 108460 kBSUnreclaim: 97628 kBKernelStack: 6768 kBPageTables: 89756 kBNFS_Unstable: 0 kBBounce: 0 kBWritebackTmp: 0 kBCommitLimit: 8233480 kBCommitted_AS: 15552704 kBVmallocTotal: 34359738367 kBVmallocUsed: 302668 kBVmallocChunk: 34359372536 kBHardwareCorrupted: 0 kBAnonHugePages: 11448320 kBHugePages_Total: 0HugePages_Free: 0HugePages_Rsvd: 0HugePages_Surp: 0Hugepagesize: 2048 kBDirectMap4k: 10240 kBDirectMap2M: 16766976 kB
Thanks.
Thanks rvillarin.Seems like we would have to dig more into the cluster status and the processes using up the memory. It is best for you to open a case with our support team to look into this.As of now, you have around 5% free memory on this CVM which is a bit of a worry, although that would not mean that something is going to breakdown.
Enter your E-mail address. We'll send you an e-mail with instructions to reset your password.