VMware Cloud Community
Darktan
Contributor
Contributor
Jump to solution

Adding multiple hosts to shared storage

Greetings....

I apologize if this has been answered. It's late and I can't seem to figure it out or find anything in my searches through the forums. It being late and me being tired means I might have missed something.

I have 2 PE1950's running ESXi and both attached to the same LUN on our SAN. One of the 1950's was created with the wrong RAID config and has to be recreated. This is fine since it's just a backup of the other host that's really doing all the work. The problem I'm running into is after I reinstalled ESXi to the system I need to reattach the SAN space. When I click 'Add Storage' and go through the motions, it wants to reformat the storage. This of course is a bad thing. Am I missing something? Do I just re-scan and the storage will automatically attach? Or do I need to do something else?

Any help would be greatly appreciative...

FYI... I'm new to the Linux side of this and would really have prefered work paying for ESX with all the tools that I'm sure would make this sooo much eaiser.

Peace

--Darktan

Tags (4)
0 Kudos
1 Solution

Accepted Solutions
weinstein5
Immortal
Immortal
Jump to solution

Welcome to the forums - if the storage is already been formatted as a VMFS datastore than your should simply have to bring up the new install of ESX and rescan the SAN and it should be available for use - you will not need to readd it - does it show up in your list of available datastores?

If you find this or any other answer useful please consider awarding points by marking the answer correct or helpful

View solution in original post

0 Kudos
11 Replies
IB_IT
Expert
Expert
Jump to solution

when reinstalling ESX, unplug your SAN fiber and reattach when finished. Then from the VI client pointed to that host, rescan all HBA's (better to do one at a time) and see what happens. I take it this was all working and both hosts could see all storage before?

0 Kudos
weinstein5
Immortal
Immortal
Jump to solution

Welcome to the forums - if the storage is already been formatted as a VMFS datastore than your should simply have to bring up the new install of ESX and rescan the SAN and it should be available for use - you will not need to readd it - does it show up in your list of available datastores?

If you find this or any other answer useful please consider awarding points by marking the answer correct or helpful
0 Kudos
langadi
Contributor
Contributor
Jump to solution

If you can access the LUN on other 1950 then re-scan should attach the LUN automaticlally in the newly installed ESXi machine.

-langadi

Darktan
Contributor
Contributor
Jump to solution

Thank you everyone.

Rescan is what I was looking for. I chalk it up to it being 1am when I was trying to shuffle these machines around and not noticing or paying attention to my options. This is first time I've added hosts to a live environment too. During our tests we just set everything up from scratch.

Hopefully this will be the last problem I run into. Yeah, probably not.... Smiley Happy

Peace

0 Kudos
dwcrist
Enthusiast
Enthusiast
Jump to solution

I have something similar. I have just added a new ESXi host. I had my SAN guy let it see the two VMFS3 volumes already seen by an existing ESX (non-i) host. I did the rescan and can see the two LUNs on both HBAs under 'Storage Adapters'. But when I go to 'Storage' there is nothing listed. Clicking on 'Refresh' does nothing. As the above poster, when I started to 'Add Storage' it warns that it is going to destroy all of the data, so I of course click on Cancel.

The existing host is 3.5.0 (82663). The new ESXi host is build level 111522. Thanks for any help.

-Doug

0 Kudos
dwcrist
Enthusiast
Enthusiast
Jump to solution

I decided to call support and he knew what it was right away. He had me do the following:

1. Bring up Advanced Settings at the bottom of the Configuration screen.

2. Highlight LVM

3. Change LVM.DisallowSnapshotLUN from 1 to 0.

4. Rescan both of the HBAs, only checking for volumes (bottom check box).

Now the existing volumes are showing up under 'Storage'.

I hope this might help someone else.

-Doug

0 Kudos
Dave_Mishchenko
Immortal
Immortal
Jump to solution

If the SAN is properly setup to present the LUN to all hosts with the same LUN id, then all you have to do is go to Configuration \ Storage Adapters and click Rescan. Disallowsnapshot would be used if for example you were presenting the LUN to a DR ESX host. See page 114/115 - http://www.vmware.com/pdf/vi3_35/esx_3/r35u2/vi3_35_25_u2_san_cfg.pdf.

0 Kudos
dwcrist
Enthusiast
Enthusiast
Jump to solution

Dave,

Yes, that's the way it's always worked for me in the past. But this is our first ESXi server and the support technician knew immediately that this would fix the problem, and it did. He also said that this is peculiar to ESXi.

Now, if I can just figure out how to Vmotion between my Intel 7350 and 7450 processors! But that's another story and I will do a search and open another thread if necessary.

Thanks,

Doug

0 Kudos
bernd2nd
Contributor
Contributor
Jump to solution

Have you resolved the VMotion issue between Intel X7350 and E7450 cpus?

I'm investigating on a similar issue. I have advised to use EVC (enhanced VMotion compatibility). But it seems not to work for some reasons.

0 Kudos
dwcrist
Enthusiast
Enthusiast
Jump to solution

Yes, I too have set up a new cluster with EVC turned on. But before that, I had already opened another SR and VMware had put some XML code into my vpxd.cfg file (in C:\Documents and Settings\All Users\Application Data\VMware\VMware VirtualCenter). I will copy the entire file below, with the caveat that this works for me, but should not be copied blindly without calling VMware support. I have bolded the section added by support.

<config>

<!-- <ws1x> -->

<!-- Enables 1.x Web Services compatibility -->

<!-- <enabled>true</enabled> -->

<!-- Enables 1.x Built in Perf data Refresh -->

<!-- <perfRefreshEnabled>false</perfRefreshEnabled> -->

<!-- Location of ws1x persistent data. Default is same directory as vpxd.cfg -->

<!-- <dataFile> C:\ws1x.xml </dataFile> -->

<!-- Location of the ws1x event mapping file. Default is the same directory as vpxd.cfg -->

<!-- <eventMap> C:\ws1xEventMap.xml </eventMap> -->

<!-- </ws1x> -->

<vpxd>

<guestOSDescriptor>

  • <esx-3-5-0>*

  • <all-versions>*

  • <all-guests>*

  • <cpuFeatureMask>*

  • <default-vendor>*

  • <level-1>*

  • <ecx>--:::xxx-::x-:---x:-x-x</ecx>*

  • </level-1>*

  • </default-vendor>*

  • </cpuFeatureMask>*

  • </all-guests>*

  • </all-versions>*

  • </esx-3-5-0>*

  • </guestOSDescriptor>*

<das>

<serializeadds>true</serializeadds>

<slotMemMinMB>256</slotMemMinMB>

<slotCpuMinMHz>256</slotCpuMinMHz>

</das>

<filterOverheadLimitIssues>true</filterOverheadLimitIssues>

</vpxd>

<vmacore>

<threadPool>

<TaskMax>30</TaskMax>

</threadPool>

<http>

<readTimeoutMs>45000</readTimeoutMs>

<writeTimeoutMs>45000</writeTimeoutMs>

<blacklistPeriodMs>3000</blacklistPeriodMs>

</http>

<ssl>

<handshakeTimeoutMs>30000</handshakeTimeoutMs>

</ssl>

</vmacore>

</config>

Anyway, I can now Vmotion between my Intel 7350 and 7450 processors.

HTH,

Doug

0 Kudos
bernd2nd
Contributor
Contributor
Jump to solution

Thank you for your quick response. We are on VC 2.5U3/ESX 3.5U3. Unfortunatly there is no vpxd.cfg on the hole VC server.

I think i will log a new SR.

0 Kudos