VMware Cloud Community
Sam_CN
Contributor
Contributor

Datastore"datastore2" is not accessible in read-write modde from host"hhesxs01"

At first, the VM HHS11 was running on the host "hhesxs01" and the datastorage is LUN3,  I have configured replicate VM to "datastore2" which was a datastorages of another host "hhesxs03". replication working fine. and I can recover the VM successfully. so the VM was running on  the host "hhesxs03" .Then I reconfigured replicate VM to LUN3, replication working fine too. But when try to recover VM to different host than source host "hhesxs01" its give following error:

Datastore"datastore2" is not accessible in read-write modde from host"hhesxs01"

I have tried to  restart the vsphere replication appliance and the host "hhesxs03", but after those operations, the problem still can not solve.

Tags (1)
Reply
0 Kudos
8 Replies
a_p_
Leadership
Leadership

Welcome to the Community,

does the VM have a reference to "datastore2", e.g. an .iso image configured for the VM's virtual CD-ROM drive?

André

Reply
0 Kudos
Sam_CN
Contributor
Contributor

Hi André,

There was not any iso image configured for the VM's virtual CD-ROM drive or other special configuration on the "datastore2". It just uses for VM's storage and vSphere Replication.

Moreover, please note the, the datastore2 and was host on different Cluster with LUN3.

datastore2.jpg

Reply
0 Kudos
Sam_CN
Contributor
Contributor

André,

The warning screen capture for you reference..

Recover warning.jpg

Reply
0 Kudos
linotelera
Hot Shot
Hot Shot

Have you tried to restart vsphere replication appliance?

Reply
0 Kudos
Sam_CN
Contributor
Contributor

Hi Sir,

Yes, I have read the following article: https://communities.vmware.com/thread/441385

so, I have tried to restart the vsphere replication appliance,and restart the Host, but the problem still existed. I do't have any idea for this.

Reply
0 Kudos
linotelera
Hot Shot
Hot Shot

Trying to redraw your scenario:

hhesxs01, hhesxs01 -> LUN3 (shared)

hhesxs03 -> datastore2 (shared or local? assume local)

VMx -> running from hhesxs01 under lun3 and replicated in datastore2.

You want to recover this VM from hhesxs01 under datastore2, but the only option to start your VM is hhesxs03 under datastore2... Is it correct?

As a first look, the error message means that you have a local datastore or datastore2 isn't in the same storage group (not cluster) of hhesxs01... I think, because VR can recover only to host which have direct access to the datastore destination, the wizard shows the read/write error. We must check san configurations/issues.

Regards

Lino

Reply
0 Kudos
Sam_CN
Contributor
Contributor

Trying to redraw your scenario:

hhesxs01, hhesxs01 -> LUN3 (shared)

>> Yes, LUN3 is a shared datastorage.

hhesxs03 -> datastore2 (shared or local? assume local)

>>datastore2 is a local datastorage for hhesx03.

VMx -> running from hhesxs01 under lun3 and replicated in datastore2.

You want to recover this VM from hhesxs01 under datastore2, but the only option to start your VM is hhesxs03 under datastore2... Is it correct?

>> Correct, The VR run fine, recover the VM from hhesxs03 under LUN3 is ok. On the contrary, when recover this VM from hhesxs01 under datastore2, It goes wrong with the above error.

As a first look, the error message means that you have a local datastore or datastore2 isn't in the same storage group (not cluster) of hhesxs01... I think, because VR can recover only to host which have direct access to the datastore destination, the wizard shows the read/write error. We must check san configurations/issues.

>> I'm confused, hhesxs01 can access datastore2 by read/write model if VR run fine , but why goes wrong on recovery?

Reply
0 Kudos
linotelera
Hot Shot
Hot Shot

>> Correct, The VR run fine, recover the VM from hhesxs03 under LUN3 is ok. On the contrary, when recover this VM from hhesxs01 under datastore2, It goes wrong with the above error.

Is LUN3 accessible from hhesxs03P?


>> I'm confused, hhesxs01 can access datastore2 by read/write model if VR run fine , but why goes wrong on recovery?

Replication walk through host simple using a vmKernel port (not by storage connections) :

LUN3 -> vmkernel on hhesxs01 -> vmkernel on hhesxs03 -> datastore2

But when you try to start VM only the host which could be read/write under destination datastore is able to bring this workload. You can replicate everywhere, but you can recover only in the host which can bring destination datastore.

Read this post vSphere replication implementation ... hope this could help