Quantcast
Viewing all 2572 articles
Browse latest View live
↧

Some VM doesn't start in test mode at the recovery site

July 20, 2016, 4:34 pm
$
0
0

Hi,

 

After running a recovery plan in test mode i can see in the report that not all VM have start on the recovery site.

In my last test my recovery plan contained 15 VMs and 2 of them (srv-01 and srv-02) haven't start on the recovery sute. One VM have vmware tools up to date and the other not so i don't think that vmware tools are the problem.

 

So that is the summary of the rapport, i can notice that an operation expired during the step "create a snapshot of writable storage" and that finish by an inactive status during the startup-up step of these VMs.

               

 

4. Créer un snapshot du stockage inscriptibleErreur - Opération expirée : 900 secondes
4.16. srv-01Erreur - Opération expirée : 900 secondes
4.16.1. Configurer le stockageErreur - Opération expirée : 900 secondes
4.17. srv-02Erreur - Opération expirée : 900 secondes
4.17.1. Configurer le stockageErreur - Opération expirée : 900 secondes
8.12. srv-01Inactif
8.12.1. Démarrage du clientInactif
8.12.2. Personnaliser l'IPInactif
8.12.3. Arrêt du clientInactif
8.12.4. Mettre sous tensionInactif
8.12.5. Attendre VMware ToolsInactif
8.13. srv-02Inactif
8.13.1. Démarrage du clientInactif
8.13.2. Personnaliser l'IPInactif
8.13.3. Arrêt du clientInactif
8.13.4. Mettre sous tensionInactif
8.13.5. Attendre VMware ToolsInactif

 

Any suggestion how to fix it ?

 

Regards.

Image may be NSFW.
Clik here to view.
↧
Search

Site Recovery Manager inquiries

July 27, 2016, 12:36 am
$
0
0

1. If I am using an 1 SRM license for both PROD and DR site, can I do failback and reprotect?

2. Client will do a network split to simulate a DR scenario (PROD will still be running), In this scenario can I do reprotect and failback testing? They dont want to just use the test recovery plan feature of SRM.

3. SRM can be used in network split (testing) or in the event vCenter Server is corrupted or vCenter manage services encountered a problem?

↧
↧

protect VM through powercli not working: Not authenticated into SRM

July 28, 2016, 1:33 am
$
0
0

We are trying to automate the protection of a VM through powercli

We followed the several posts available on internet like fe http://www.automate-it.today/automating-srm-powercli/

However we end up with the error Not authenticated into SRM <remote vcenter server name>

 

What's not working in our enviromnemt is to log in with -RemoteUser or -RemoteCredential to SRM although the user specified has the priviliges to log in.

 

Anyone a suggestion how to solve this?

Image may be NSFW.
Clik here to view.
↧

Powering of VMs during failover

July 31, 2016, 3:06 pm
$
0
0

Hi!

 

If there are non-protected VMs running in a DR site and a recovery action is initiated from PROD to DR, will SRM prioritize to power on the configured protected VMs and will shutdown the non-protected VMs to accommodate resources?

 

What if there are enough resources in the DR site?

What if the resources in the DR site is enough to accommodate the protected VMs?

Image may be NSFW.
Clik here to view.
↧

SRM recovery test scenario

August 1, 2016, 1:39 am
$
0
0

One of our client wants to test the SRM capability during recovery, can these scenario be possible to simulate a recovery plan?

 

1. Use the test recovery plan in SRM. After test replica VM is powered on, is it possible to test the functionality of the test replica VM in the DR site? Test connectivity, application functionality? Then after this, cleanup will be initiated.

 

2. Split the network between PROD and DR, use the recovery plan (not test) in DR network then test the VMs functionality? My concern if after testing and the network is reconnected (DR VMs should be powered off this time), what will happen to the replication of the VMs from PROD to DR, will these cause replication corruption?

 

3. Power off VM/s to be tested and vCenter Server in PROD and then use recovery in DR site to actual test the VMs. My issue here if this can work and if corruption will occur once the VMs is powered on in PROD (replication, DR VMs already powered off)

 

The client doesn't want to fully power off their PROD site since not all VMs are protected for DR but still are crtical VMs.. They just want to test SRM functionality and not just use the test recovery feature of SRM.

↧
↧

Test Scenario using vSphere replication

August 1, 2016, 8:48 am
$
0
0

We have a client of ours who are running SRM 5.8 with vSphere Replication (No san based replication)

Test failovers will not work in this test scenatio

 

They would like to do a planned failover to the recovery site and do some tests on the recovered VM's after which they would like to do a failback to what was the protected site

Simple question really when we do a failback is there an advanced config option in SRM that says we do not want to synch back any changes made to the VM's we had failed over.

So in this scenario we can take advantage of the SRM workflows do a planned failover and then when we do a failback we do not want to replicate any changes back to what was our old protected site

Hope that helps

 

Thanks in advance

Image may be NSFW.
Clik here to view.
↧

SRM Attributes

August 3, 2016, 7:35 am
$
0
0

My customer is trying to ascertain what attributes for virtual machines are transferred across during a SRM failover as reservations still don't and they have the reservations to stop the creation of the vmswap files due to their tight storage allocations (tried to fix this but so far they are refusing) Is there a list of these attributes?

↧

SRM Fail-over test - Pulling the plug

August 3, 2016, 1:36 pm
$
0
0

For our test want plan to “pull the plug” (figuratively) on the production data center prior to performing the fail-over test.  We want to simulate an actual data center disaster.  All networks, routing, firewall, outside NAT's, etc. will be failed over to the DR site.  Our goal is to not only test recover ability of servers but also outside access to inside resources.

 

At the DR site I would like to launch vCenter and then in SRM click the TEST link.

 

However - The TEST function will not work unless the production and DR vCenter servers remain in constant communication.

 

Question – Would it work if I moved the vCenter server from the production site to the DR site prior to pulling the plug?  Then both vCenter servers will remain in continuous communication.  The vCenter server database is local.

 

Regards,

Terry

↧
Search

Site Recovery Manager Failover issue

August 5, 2016, 7:40 am
$
0
0

I would like to consult an issue we encountered during test recovery.

We encountered "VMware Tools timeout error in ...... seconds"

Tried to increase value but still problem exists.

My question is:

 

1. Can I just uncheck the VMware Tools to wait? When I do that, test recovery plan works. Will there be any implications if I disable that and an actual failover really occurred?

2. If I have 10 VMs and one VM 1 is a dependency of VM 2 before it can start, If an error occured on VM 1, will VM 2 bypass that? I assumed this is for dependencies configured but in a scenario where 10 VMs with VM 1 is priority 1 and VM 2 is priority 2, I assume it will work?

3. VM encountered with error is running in VM Tools 9 (windows 2008). Will this VM tools is compatible with the latest SRM version?

Image may be NSFW.
Clik here to view.
↧
↧

Inquiries

August 6, 2016, 10:09 am
$
0
0

Hi,

 

I would like to ask a few questions:

 

1. Can we permanently set the account recognition for SRM for PROD and DR? Not having to enter the credentials after browser log off?

2. If recovery and cleanup stucks to to N%, can we force stop the activity?

Image may be NSFW.
Clik here to view.
↧

VMware SRM/SRA Certification question about SetupiSCSITests

August 7, 2016, 11:29 pm
$
0
0

Hi,

In the beginning, we have first certification test case failed at SetupiSCSITests,

 

 

We use workbench to run SRA certification

the first item: SetupiSCSITests

Workbench show ERROR as follow:

 

[testng] ,123 INFO  [main] - 'VlsiClientBase.resetClient():210' Shutting down client connection for: https://192.168.139.171:443/sdk

INFO:      [java]    [testng] ,204 INFO  [main] - 'VimClient.login():309' Connection: 192.168.139.171:443, Username: administrator@vsphere.local, Password: P@ssw0rd

INFO:      [java]    [testng] ,205 DEBUG [main] - 'NetworkConnectionUtil.waitForConnection():74' Waiting for 192.168.139.171 to be listening on port 443.

INFO:      [java]    [testng] ,206 INFO  [main] - 'TaskMonitorUtil.waitForCustomTask():156' Waiting for custom task for 5

INFO:      [java]    [testng] ,208 DEBUG [main] - 'NetworkConnectionUtil.waitForConnection():84' 192.168.139.171 is now listening on port 443.

INFO:      [java]    [testng] ,585 DEBUG [main] - 'DrClient.login():424' Logging into DR session manager

INFO:      [java]    [testng] ,697 ERROR [main] - 'DrTestBase.beforeMethod():110' Unable to login into clients or remote sites.

INFO:      [java]    [testng] java.lang.RuntimeException: com.vmware.srm.main.core.test.TestException: Unable to login

 

 

Looks like it can not login my Vcenter (my vcenter IP is 192.168.139.171)

1. what is DR session manager in thiscertification test case ??

2. I very sure the vcenter SSO account and password is correct what we input on workbench parameter

3. Maybe miss something when we develop SRA or setting SRMenvironment??

 

Anyone have idea??

 

Thank you very much,

Nicole

Image may be NSFW.
Clik here to view.
↧

host lost access to lun (datastore) after planned failover test

August 8, 2016, 4:02 am
$
0
0

Hello,

I have the following config.

A production site with vsphere 5.5, SRM 5.5 and an ibm storage array (Storwize V3700),

A Recovery site with the same config,

The two storage arrays are connected via FC and an asynchronous replication is running.

We did a planned failover with SRM, everything run smoothly, but at the end of the test (while all VMs were UP in the DR site) i noticed that some datastores disappeared from some hosts.

After restarting the hosts the LUNs appeared correctely. But i noticed the same behaviour while failing back to the primary production site.

 

Thanks in advance for your help.

↧

inventory mappings

August 9, 2016, 12:10 am
$
0
0

I would like to know why I can have only 1 available network mapping for reverse mapping?

What if I have 3 separate networks in the PROD site that has also a 3 network counterpart in the DR site?

Will one of the 3 network mappings will be enabled?

Image may be NSFW.
Clik here to view.
↧
↧

SRM for protecting other hypervisors VM and non vmware cloud vms

August 9, 2016, 8:35 am
$
0
0

As per my understanding srm cannot be used to protect other hypervisors VM and non vmware cloud vms, Correct me if i am wrong? Can it be used to protect openstack vms?

↧

Site Recovery

August 9, 2016, 10:27 pm
$
0
0

Lets say in my PROD, i have the following:

 

App1, Web1, DB1, App2, Web2, DB2

 

I created 3 recovery Plan:

 

Recovery Plan1: App1, Web1, DB1, App2, Web2, DB2

Recovery Plan 2: App1, Web1, DB1

Recovery Plan 3: App2, Web2, DB2

 

1. If I triggered Recovery Plan 1 and DB1 did not powered on, Can I just manually power on the replicated DB1 VM in DR?

 

2. Follow up on no. 1, Can I also trigger Recovery Plan 2 if DB1 in Recovery Plan 1 failed to start? What would be the implications?

 

SRM can do Disaster Avoidance and Datacenter migration right?

If i triggered the failover/recovery button while PROD is running.....

 

3. In recovery Plan 1, all VMs on PROD will shut down and recovered in DR

 

4. In Recovery Plan 2 for example, If I triggered failover/recovery; only VMs from this recovery plan will be shut down in PROD and powered in DR? or it will still power off the whole PROD site?

Image may be NSFW.
Clik here to view.
↧
Search

VM's in Protection Groups shows invalid message : invalid: com.vmware.vcDr.dr.replication.fault.vmprotectionlost.summary

August 10, 2016, 12:05 pm
$
0
0

After failover, our LUN replicated as snap. We renamed it by removing snap prefix. Then later our protection group shower warning

There are configuration issues with associated vm's

Datastore "****" is not replicated.

 

On vm's in protection groups is says following error

invalid:com.vmware.vcDr.dr.replication.fault.vmprotectionlost.summary

cause:com.vmware.vcDr.dr.storageprovider.fault.configgilenotreplicated.summary

 

Any idea how to fix this ?

↧

SRM with Multiple Storage Arrays and Array based Replication

March 1, 2016, 11:53 pm
$
0
0

I have few queries for which I am finding answers.

 

My Environment:

Protected Site

vCenter 5.5

SRM 5.8

EMC Recovery Point for Replication

EMC VNX as Storage Array

EMC SRA configured in SRM to talk to EMC Recovery Point

 

Recovery Site

vCenter 5.5

SRM 5.8

EMC Recovery Point for Replication

EMC VNX as Storage Array.

EMC SRA configured in SRM to talk to EMC Recovery Point

 

I have a cloud solution running, as I onboard a customer.

- Customer is a resource group on vCenter.

- Datastore Cluster for that Customer with set of Datastore as per the Quota is also provisioned

- Same set of provisioning is performed automatically at Recovery Site.

- Dedicated Protection Group and Recovery Plan is created for the Customer with attached Datastore Cluster in SRM configuration.

 

This is happening normally with EMC VNX as Storage Array.

 

We are planning to introduce another Storage Array (all flash/high IOPs) which will be integrated with EMC Recovery Point as well.

 

I want to provide the customer both options together for which I am confused how SRM will behave as SRA will detect LUNs from both Arrays which will be visible to SRM.

The datastore cluster will have mixture of datastores for different array.

 

Just wondering if there is any challenge in the following situation or it will be straight forward? as for myself all I want is a replicated LUN/Datastore from Storage and to be visible by SRM through SRA adopter.

↧
↧

SRM and only 1 SAN

August 14, 2016, 2:04 pm
$
0
0

Hi!, I have the following situation:

 

Two sites, the primary and the recovery site.

In our primary site there's only 1 SAN powered by a humble Hitachi storage appliance, and this replicates on the recovery site.

 

Our SRM was installed, as you con guess, next to all of our other VMs, on the same storage appliance. What I do know is that if our storage goes down, SRM will go down with all the other VMs and we are going to be unable to actívate any recovery plan.

 

Could you please tell me if this is the recommended deployment for SRM? or should we install SRM on another physical server to protect our DRP?.

 

I mean, the storage goes down, but still our live SRM could tell to our recovery site to power on and take control.

 

Am I right?.

 

Thank you very much in advance for your time and your knowledge!.

Image may be NSFW.
Clik here to view.
↧

SRM inventory mappings

August 15, 2016, 9:25 am
$
0
0

What will happen to PROD VMs during failover/recovery if they were moved from the configured inventory mappings?

example: used a different network or storage or cluster

↧

NetApp SVM and Protection Groups

August 15, 2016, 5:07 pm
$
0
0

Hi. I have a customer that has created a single NetApp SVM (storage virtual machine) that houses all of their Primary Site datastores (i.e. multiple NetApp volumes/luns that map to multiple datastores). We've successfully created the Array Manager and see the local and remote array pair. We can also see all of the multiple datastores and the replication direction.

When we create a Protection Group, all of those datastores show up as a single Datastore Group. So, we only have the option of creating one giant Protection Group that has all datatstores. This is not what we want. We were hoping that the datastores would each show up as separate Datastore Groups, so we could create multiple Protection groups.

My question is, when SRM computes Datastore Groups, does it look at a NetApp SVM as the smallest unit of granularity? Meaning if there's only a single SVM on the NetApp storage, will SRM treat all datastores within that SVM as a single Datastore Group/Protection Group?

Ultimately, we'd like the ability to define Protection Groups that might only have a single datastore, or maybe house a couple of datastores. We're just a little confused on how SRM treats NetApp SVM's in terms of computing SRM Datastore Groups.

 

Thanks

↧
Viewing all 2572 articles
Browse latest View live
Search
<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>