Link to home
Start Free TrialLog in
Avatar of philb19
philb19

asked on

Xangaiti not picking up datastores to monitor

Hi,

Urgent help on xangati - Not picking up datasatores to monitor - odd 1 maybe bug. But I cant performance monitor on the hybrid volumes attached to vcenter 5.1 - the only datastore being picked up is the sata datastore. - I cant drill down into at any granular level any other datastore?

Netapp cluster mode release 8.3 - xangaiti only monitoring the 1 (least used sata data-stores). Vcenter 5.1 - could be esx bug vcenter

Hopefully Andrew catches this question :)
Avatar of Andrew Hancock (VMware vExpert PRO / EE Fellow/British Beekeeper)
Andrew Hancock (VMware vExpert PRO / EE Fellow/British Beekeeper)
Flag of United Kingdom of Great Britain and Northern Ireland image

So Xangati is not detecting any shared network storage ?

just local storage ?

and I assume that all these shared storage volumes, are connected to all hosts ?

Are these iSCSI or NFS ?

What version of vCenter Server and ESXi 5.1 ? latest versions ?

If they are seen by vCenter Server they should be picked up by Xangati?

You have refresh and re-scanned the datastores ?

or a bug with Xangati
Avatar of philb19
philb19

ASKER

Thanks Andrew

No the SATA is shared storage - NFS  - there is no iscsi at play here all NFS
Vcenter 5.1.0 build 1473063
no local storage - all netapp
connected to all hosts Yes
esx 5.1.0 version 2000251
refresh tried yes.
Avatar of philb19

ASKER

its the hybrid Netapp  sas ssd datastores that cant be seen

attached to all hosts in prod cluster (esx)
That seems very odd, as at that layer, ESXi cannot see the underlying storage, it's all seen as NFS.

does it look any different, in the way it's presented to ESXi ?
Avatar of philb19

ASKER

Hi Andrew - This is cluster mode Netapp - I was more familiar with the NFS export process on 7mode which changed somewhat in cluster-mode.
But I will have a look at more closely tomorrow based on your question,

I have another question - let me know if you require me to place in a new question.
There is 1 VM OS win2008 32bit (not r2)  that has an e1000 driver for it vNIC  - so not using VMXNET driver. I realise that the vmxnet driver provides better performance. Im just wondering if the performance loss is significant using e1000 and is likely to effect application performance significantly?

The reason a e1000 driver was used is that the vm was put in early days running esx5 or 4.1 - at the time I recall that we couldn't get the vmware nic driver to work in the OS so we fell back to e1000 that worked everytime!
Thankyou for help
SOLUTION
Avatar of Andrew Hancock (VMware vExpert PRO / EE Fellow/British Beekeeper)
Andrew Hancock (VMware vExpert PRO / EE Fellow/British Beekeeper)
Flag of United Kingdom of Great Britain and Northern Ireland image

Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial
Avatar of philb19

ASKER

Thanks Andrew Awesome as always :)

I did pickup a difference in the presentation of the datastores:

the SATA nfs which xangati can get monitoring from is presented over a different VIF ( IP = .6) to the esx hosts
While the Hybrid ssd/sas nfs datastores are presented over the other vif (IP = .7) to the esx hosts

Now while all datastores are of course all mounted on all of the esx hosts on the cluster - this is a difference.

In fact around a month ago - we couldn't nfs mount using 1 of the vifs from a linux server. Just by trial and error we tried going to the other vif and it did nfs mount ok. We then went straight back to the other vif and suddenly it mounted ok as well. - Some weird netapp issue maybe.

Our external guy seems certain that the xangati issue is a bug in the vcenter version.

any further ideas? Cheers
ASKER CERTIFIED SOLUTION
Link to home
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Start Free Trial