VMware Cloud Community
Prost
Contributor
Contributor
Jump to solution

Alarm 'Cannot connect to storage'

I am running ESXi 4.0 Build 261974 (this is an HP version of ESXi and has Systems Insite Manager Imbeded) and vCenter Server 4.0.0 Build 208111

5 of my servers share a EMC CX3 in Site 1 and I have 1 host in Site 2 on connected to an EMC CX3.

I have been running on vSphere for a couple of month and everything is running fine. This morning I was reviewing the event log (I have not reviewed them since the build) and noticed 2 alarms comming from 2 of my 6 hosts each are from different sites. I am getting these every 5 minutes and it filled up my log.

Alarm ' Cannot connect to storage' on entity triggered an action.

I can connect to the storage the VM Guests are all fine.

Can someone tell me what these are from?

Reply
0 Kudos
1 Solution

Accepted Solutions
Silverchenau
Enthusiast
Enthusiast
Jump to solution

Well. First of all, I would say, Unless you are using VMFS 2, there are no major difference in VMFS 3.x. It should be all compatiable.

If you are using VMFS 2, in ESXi 4.1(not quite sure about 4.0), your VMFS will be read only according to VMWARE docs.

I think I had similar issue before with my ESX 3.5, but it happened only once and I think you can fixed it by reinstall ESXi since it's not difficult procedure.

Just make a backup and reinstall and restore your configuration.

- Silver

My Vmware blog:

- Silver My Vmware blog: http://geeksilverblog.com

View solution in original post

Reply
0 Kudos
12 Replies
Silverchenau
Enthusiast
Enthusiast
Jump to solution

It appears you have misconfigure LUNs. The possible reason could be you have masked one LUN from SAN but your ESX server hasn't manually scan the datastore info.

Suggestion:

run rescan on all your storages






- Silver

My Vmware blog: http://geeksilverblog.com

- Silver My Vmware blog: http://geeksilverblog.com
Reply
0 Kudos
Prost
Contributor
Contributor
Jump to solution

Thanks I just ran “Scan for New Storage Devices” and Scan for new VMFS Volumes” on the host that is sending the alarm in Site 1 and the host that is sending the alarm in Site 2.

I will let you know the result

Last alarm was at 7/29/2010 6:59:43 AM

I will keep you posted.

Reply
0 Kudos
Prost
Contributor
Contributor
Jump to solution

Nope just got more alerts

Reply
0 Kudos
Prost
Contributor
Contributor
Jump to solution

I have had some trouble in the past with the persistent reservations holding a lock on my LUNs. I am going to try and trespass my LUNs to free any locks.

Reply
0 Kudos
Prost
Contributor
Contributor
Jump to solution

still happening Smiley Sad

Is there a way to get more information like which HBA is causing the issue or somthing?

I have attached a view of the alarms that I am looking at.

Reply
0 Kudos
Silverchenau
Enthusiast
Enthusiast
Jump to solution

Maybe it's time to pull out all logs. vmkwarning, hostd and find out more details from there.






- Silver

My Vmware blog: http://geeksilverblog.com

- Silver My Vmware blog: http://geeksilverblog.com
Reply
0 Kudos
Prost
Contributor
Contributor
Jump to solution

Excuse my ignorance but where are these logs? Can I view them from vCenter or do I need to open the host console and enter the “unsupported” mode. (I am running ESXi)

Thanks,

Reply
0 Kudos
Silverchenau
Enthusiast
Enthusiast
Jump to solution

All right. I see you need a little bit more work on logs.

But for now, you can use vCenter to export that server log. It will run vm-support on the esxi and download all logs to your vCenter server and compress them into 2 files. You need to unzip the files and find vmkwarning, vmkernel (cross check the time with time you have issue) and hostd as well.

If you need more help, pls go to http://geeksilver.wordpress.com/2010/06/16/esxi-era-using-esxi-to-replace-esx-%e2%80%93-log-files-pa...

This post is for 4.0 but it will help you little bit more in the future.






- Silver

My Vmware blog: http://geeksilverblog.com

- Silver My Vmware blog: http://geeksilverblog.com
Reply
0 Kudos
Prost
Contributor
Contributor
Jump to solution

All I can say is WOW.

Thanks for the information I reviewed the link you sent and am very impressed with your blog. I have recently migrated to ESXi when I upgraded to vSphere and have been struggling with not having access to the console. Yes I have found the unsupported command but if VMware doesn’t support it I would have thought they would have given you a different path to get the same tasks done. Anyway keep up the great work on your blog.

I am going to need some time to get my arms around the logging so I will get back to you next week.

Reply
0 Kudos
Silverchenau
Enthusiast
Enthusiast
Jump to solution

hi, Prost:

ESXi is not long "unsupported" for 4.1 version and later on. Just let you know about it. Smiley Wink






- Silver

My Vmware blog: http://geeksilverblog.com

- Silver My Vmware blog: http://geeksilverblog.com
Reply
0 Kudos
Prost
Contributor
Contributor
Jump to solution

OK I have contacted VM-Support and uploaded the logs to have them reviewed. They say one of my hosts has a LOCK on the datastore and is not releasing it. Yesterday this happened for an hour and I lost access to some of my VMs.

I asked support when I upgraded if I needed to upgrade the VMFS and they said NO. Is this true because it looks like it is only happening to my old VMFS datastores?

Reply
0 Kudos
Silverchenau
Enthusiast
Enthusiast
Jump to solution

Well. First of all, I would say, Unless you are using VMFS 2, there are no major difference in VMFS 3.x. It should be all compatiable.

If you are using VMFS 2, in ESXi 4.1(not quite sure about 4.0), your VMFS will be read only according to VMWARE docs.

I think I had similar issue before with my ESX 3.5, but it happened only once and I think you can fixed it by reinstall ESXi since it's not difficult procedure.

Just make a backup and reinstall and restore your configuration.

- Silver

My Vmware blog:

- Silver My Vmware blog: http://geeksilverblog.com
Reply
0 Kudos