Lost access to volume

[dbsid]

Hi all,

I have a Ml350 gen9 with a smart array b140i, bought on april 2015, installed end of april.

After a few months of working with esxi 6.0 (HP custome image) today we are expecting a lot of problem with the virtual machine running of that server.

In the event log of the vsphere client, we using esxi we are very small company, in the log we can see

continuosly error like :

Lost access to volume ... due a connectivity issue. Recovery attemp is in progress

after few second or 1 minute depends

Successfully restored access to volume

This gave us a lot of problem because the virtual machine seems that are running but their are freezed so no login to the domain is permitted and the other service doesn't works.

Please I need some help!

[Renmo]

The first thing i would recommend is to update the server's firmware, download the HP SPP and update it

[continuum]

It is interesting to see how different the answers to such a question will get depending on the role of the guy who gives his tips.

As a vSphere architect you seem to assume that fixing the connectivity issues is the logical next step ....
I have a recovery background and so my first priority would be to avoid a complete dataloss - connection issues today can be early warning of a VMFS-failure tomorrow - so I would first try to extract as many data from that datastore as long as that is still possible.
Next I would try to bring up those VMs that offer important services from a temporary location - if necessary even a Workstation host could be used for that in an emergency.

Once the most important data is extracted and the most important VMs are up and running again I then would start to look for problems with firmware,drivers or other design issues.

Ulli

[Renmo]

It's indeed an interesting point continumm.

But tell me what would you do if woke up one morning, went to your car and didn't start, which option would you choose ?

1- assume the battery is out and try to connect another car in order to start it ASAP and avoid losing working hours from your job (Firmware update)

2- call a garage to take the car and avoid touch the car to preserve it and lose working hours from your job (Recovery option)

most of HP servers run into problems when their firmware is too old.

[jredwine2857]

Hello!  Did you ever figure anything out.  I am having the same issue on a Dell PowerEdge t430 (brand new) running the Dell image of ESXi 6.  It randomly loses the datastore and also logs events stating high latency the to data store.  This is a VERY small environment and has 6 Sata drives in a RAID 6, with ESXi running from dual SDM. 

It has but a single Windows 2012R2 VM running on it with only 16 of the 48GB of RAM assigned to it.

It is ONLY a file server, no sql, no Exchange, not even printers.  It does AD/DNS/DHCP and File Sharing. And all for only 5-6 users of which only 2-3 are using it daily.  (rest are out of the office.)

Smells like either an esxi bug or something buggy with firmware on the PERC H730.

BTW.  What version of esxi are you running?  I am running ESXi 6.0 2494585.