I have seen quite a few questions come in on how to properly upgrade from an existing vCenter Server Appliance (VCSA) 5.x and/or 6.x environment to the latest vSphere 6.0 Update 1 which was just released today. Before I jump straight into the process, I think its worth covering on how updates (patches) and upgrades have traditionally been handled for the VCSA. In an update or patch scenario, you are staying within a major release of vSphere (e.g. vSphere 5.0) and moving to something like vSphere 5.0 p01 and in this case, an in-place update or patch is performed. In an upgrade scenario, where you are moving from one major release (e.g. vSphere 5.0) to another major release (e.g. vSphere 6.0), a "migration based" approach is taken. This means that you would need to deploy the new VCSA that you wish to upgrade to and then migrate the data from your old VCSA appliance to the new one which is part of the upgrade workflow. This "migration based" approach was also true for any "U" (Update) releases (e.g. vSphere 5.5 to vSphere 5.5 Update 1).
For major releases, this makes perfect sense and provides customers a nice way to easily rollback if something goes wrong. You simply power off the new VCSA and then power on your original VCSA and you are back in business. For update releases, we have heard from our customers that this process was not ideal and though there is always a risk when updating software (which is why I always recommend customers test thoroughly in a Dev/Test environment before moving to production), the amount of changes in the code is significantly less when compared to a major upgrade. One of the new features that we have introduced in vSphere 6.0 Update 1 is an in-place upgrade for "U" (Update) releases which I have already blogged about here among other new features.
This means that if you are coming from a VCSA 6.x environment and you wish to upgrade to vSphere 6.0 Update 1, you simply just mount the vSphere 6.0 Update 1 Patch ISO to your VCSA 6.x environment and perform the update from the command-line via the appliancesh interface. This is quite nice as it reduces the need to copy data between your old and new appliance and helps reduce the overall downtime. In fact, you can upgrade to vSphere 6.0 Update 1 in about 10min or so using this new method. If you are coming from a VCSA 5.x (5.0, 5.1 or 5.5) environment, this would be consider a major to major upgrade and you would need to follow the "migration based" approach to upgrade to vSphere 6.0 Update 1. One other thing to note after you have upgraded to vSphere 6.0 Update 1, we have now re-introduced URL based patching via the VAMI interface. This means in the future, you no longer need to update or patch from an ISO but can do so directly from VMware's online repository.
Below are the instructions on upgrading from VCSA 6.x to VCSA 6.0 Update 1:
Note (09/14/15):
- If you have an External PSC with your VCSA 6.x and wish to upgrade, the process shown below is the same for both the PSC and the VCSA. You will want to first upgrade your PSC first as that provides authentication to your vCenter Server. Once the PSC has been upgraded and accessible on the network again, you will then want to move to your VCSA. If you are interested in the proper sequence and ordering of VMware Products to update, you can also check out this handy VMware KB 2109760 which provides all the details
- Thanks to fellow reader Idan for reporting this but it looks like after an upgrade of the VCSA, the default VMware URL for the VAMI is not working. You will need to update it to point to the following URL https://vapp-updates.vmware.com/vai-catalog/valm/vmw/647ee3fc-e6c6-4b06-9dc2-f295d12d135c/6.0.0.10000.latest/ instead of the default one as shown in the screenshot below. This is only applicable for upgrade scenarios. If you deploy a new VCSA 6.0 Update 1, it will automatically be using the correct URL
Step 0 - Ensure you have a proper backup and take a snapshot of your VCSA 6.x appliance before beginning.
Step 1 - Download the VCSA 6.0 Update 1 Full Patch (VMware-vCenter-Server-Appliance-6.0.0.10000-3018521-patch-FP.iso) by visiting the VMware Patch Download site.
Step 2 - Mount the VCSA 6.0 Update 1 Patch ISO to your VCSA 6.x appliance using either the vSphere Web/C# Client
Step 3 - Login to your VCSA 6.x appliance via SSH to the appliancesh interface. If you have disabled that, simply type "appliancesh" and login with the root credentials.
Step 4 - Run the following command to stage and install the patches from the VCSA 6.0 Update 1 Patch ISO:
software-packages install --iso --acceptEulas
Note: If you run into any errors while either staging or installing the patches, you should drop into the bash shell and take a look at /var/log/vmware/applmgmt/software-packages.log file for additional information. One common issue that I have seen in the past is if your /storage/log partition is full and you may need to perform a clean up before continuing.
Step 5 - Once the upgrade has completed, you just need to reboot your VCSA by running the following command:
shutdown reboot -r "Updated to vSphere 6.0u1"
Step 6 - A quick way to confirm that you have successfully upgraded your VCSA to vSphere 6.0 Update 1, simply open a browser to the following URL: https://[VCSA-IP]:5480 and it should take you to the new HTML5 VAMI interface.
If you would like additional information, take a look at this VMware KB 2119924.
Now do what you did to the VAMI to the actual Vcenter. (HTML5) ☺
Hi
I make a lab with vsphere 6U1 and i have a problem with the client integration it doen't work with IE (Server 2012 & R2 & 8.1 and 10) firefox 40.0.3 idem and chrome 45 don't work broken npapi I see the vmware KB http://vmw.re/1izis03 but the problem i wasn't able to play with ovf file upload file. All my VM are in French. Do you have an idea ?
Thanks always a pleasure to read from you
Hi William, Thanks for this article.
I was wondering, is it safe to Upgrade the VCSA6 Virtual Hardware and VMware Tools ?
The reason I'm asking is because Update Manager shows these are not compliant.
My environment is vSphere 6.0U1 and VCSA 6.0 (soon to be U1).
Thanks,
Idan.
Hi Idan,
Though you could, I wouldn't recommend it, especially as these are tested at a particular configuration which could impact the system with future updates. Remember, that Virtual Hardware version maps to the capabilities of a particular version and unless you need said capability, its okay to be out of date and in fact we've tried to change the terminology few years back to Hardware Compatibility and you would only upgrade if you need to take advantage of a new feature introduced in a new vHW version. Another reason for having slightly older vHW/VMware Tools is to ensure it runs on multiple versions of vSphere, so this is also done on purpose from that perspective
Thanks 🙂
William, per VMware's upgrade PDF, there seems to be the option for doing an inplace upgrade from 5.x to 6.x.
http://www.vmware.com/files/pdf/products/vsphere/VMware-vsphere-601-vcenter-server-appliance-cmdline-deploy-and-upgrade.pdf
Any thoughts here?
What page do you see that on? 5.x is "migrate based" upgrade, which means you'll be deploying a new VCSA 6.x appliance and it'll connect to your existing appliance to copy the data across. This would not be an "in-place" upgrade
I don't see anywhere in the PDF where it differentiates updating vs upgrading processes for different versions (other than obviously the templates). Essentially, mount ISO, create template & update.
The PDF is specifically on upgrading which covers 5.x (5.1, 5,5) to 6.0u1, for patching/update this is via the Patch ISO which is what I blogged about
Am I missing something then? Because the PDF reads as if inplace upgrade is an accepted route.
Sorry if you've gotten that impression. As I said earlier, the document is specifically for upgrades (non-inplace) moving from VCSA 5.x to 6.0u1. If there's a particular page which is adding to the confusion, I can always share that back internally
I missed this sentence in the Overview section. "The upgrade of the vCenter Server Appliance is a migration of the old version to the latest version, which results in the deployment of a new vCenter Server Appliance." Seems to be the only place a new vCSA is mentioned. Even the instructions on page 5 say to take a snapshot of the old one, which seems pointless if data is just being copied out of it. Nothing mentions whether the old one is shut down at all, again implying that it's an inplace upgrade. Your blog makes this clearer than the documentation.
What if we want to upgrade from 6.0 to 6.0 Update 1, using the old method, where it migrates data to a freshly deployed appliance? Is this no longer possible? Is there some other way, either supported or unsupported, to migrate to a fresh appliance between Update releases?
That's correct, this is no longer possible and only an in-place upgrade is supported. The migration base method has pre-checks in place to even prevent it, so you wouldn't even be able to do it even if you tried.
Do we really want to reboot the appliance after the update is done, Pls confirm..
Pls ignore, Got the answer 🙂
Any update on when the bits to migrate from an embedded PSC to an external one will be available? Really the only thing holding back our v6 migrations at several clients right now as they need external PSC going forward (multi-site, multi-vCenter).
Steve,
It's available as of vSphere 6.0 update 1 🙂 You can read more about it here http://www.virtuallyghetto.com/2015/09/whats-new-in-vsphere-6-0-update-1-for-vcsa.html and there will be KB's with even more info in the coming week
Ah, so I misread the release notes then - I assume this means during the migration process and not post migration?
"vCenter Server with embedded Platform Services Controller cannot be migrated automatically to vCenter Server with external Platform Services Controller. Testing of this migration utility is not complete.
Before installing vCenter Server, determine your desired deployment option. If more than one vCenter Servers are required for replication setup, always deploy vCenter with external Platform Services Controller."
What would be the steps then to go from vCSA 5.5 w/ embedded SSO to 6.0u1 w/external PSC? It looks to me like you would need to upgrade vCSA 5.5 to 6.0u1 w/embedded, deploy a new PSC, configure PSC, then re-point vCSA to external PSC - is that right?
To clarify - do I have to start from scratch with SSO, etc. when moving from embedded to external?
Steve,
That's correct, you would go from VCSA 5.5 (Embedded) -> VCSA 6.0u1 (Embedded) then use the cmsso-util (http://pubs.vmware.com/vsphere-60/index.jsp?topic=%2Fcom.vmware.vsphere.upgrade.doc%2FGUID-E7DFB362-1875-4BCF-AB84-4F21408F87A6.html) to get to VCSA 6.0u1 w/External PSC which will require a new PSC deployment prior
No, you would not be starting from a new SSO, you would deploy a new one and setup replication with the Embedded PSC. I also did a VMworld session on this topic, in case it helps http://www.virtuallyghetto.com/2015/09/vmworld-session-vcsa-inf5975-inf4528-available-to-public.html
Hi William,
I upgraded all ESX servers and vCenter to 6.0U1 and it went smoothly.
When accessing the new VAMI and clicking on check for updates via URL, I receive an error.
VCSA has full internet access, so am I doing anything wrong?
Thanks,
Idan.
I think I found the issue, in the default repositpry, it wants to access the URL:
https://vapp-updates.vmware.com/vai-catalog/valm/vmw/647ee3fc-e6c6-4b06-9dc2-f295d12d135c/6.0.0.5110.latest/
But when I browse for it from a web-browser, it shows "File not found.""
Maybe that's the issue ?
If you deploy from the GA build, the default repo points to https://vapp-updates.vmware.com/vai-catalog/valm/vmw/647ee3fc-e6c6-4b06-9dc2-f295d12d135c/6.0.0.10000.latest/ which looks newer.
Having said that, you won't be able to open browser to the URL since its metadata behind the endpoint, so seeing "File not found" is expected.
The real test is just clicking on "Check updates via URL" and it should say, no updates.
Make sure you can fully resolve vapp-updates.vmware.com from the VCSA and that you can also ping it, it's possible you might have it blocked further upstream
When I click check for downloads I get a error
From the update server. The VSCAcan fully resolve the address.
Should I update the URL manually in the settings ?
Try updating the URL and see if that helps
That fixed it, Thanks! went to Settings, "Use Specified Repository" and pasted the link you gave me. I actually think this should have been taken care of by the VMware servers as the original link doesn't exist as this is a new feature.
So to sum things up for now, we have a fully upgraded environment except for VUM. Apparently, we need to wait 3 business days before the download of the new Module ISO is available to us (sent a support case to try to make that faster, our production support is till 2018).
Regarding upgrading VUM, do I need to do something special except for the usual shutdown services, snapshot, next, next, next ? How do we integrate to the WebClient and upgrade the C# client ?
I read the new documentation and it's a bit brief.
Thanks, Idan.
Well, I was able to download the new ISO, updated VUM to 6.0U1, then updated the plugin in the C# plugin.
*BUT* I get a nasty issue from the WebClient...
The Update Manager plugin was automatically registered during the upgrade process. VCSA6 shows that it has version 6.0.0.27278 installed.
When I open update manager, I see all records in setting blank, such as (SOAP, IP, etc). When I try to edit them, I get the following exception:
An internal error has occurred - Error #1009.
When I try to navigate around, I get the following alert in the bottom right:
"com.vmware.vim.binding.integrity.IntegrityDescription cannot be cast to com.vmware.vim.binding.integrity.IntegrityDescription"
I opened a support case, but does anyone encountered this issue ?
Thanks,
Idan.
This sounds like something I saw someone report internally. Did you restart the VCSA after the VUM upgrade? Depending on the ordering, you might have gotten into a case where the old plugin was still being loaded. The process/order of upgrade can be found in this KB http://kb.vmware.com/kb/2130672 The fix was to simply restart the vSphere Web Client service so the new plugin is loaded. If you're still having issues, I recommend filing and SR if you haven't already
Update: After rebooting the VCSA6 after updating VUM to 6.0U1 the vSphere server and VUM play together.
New issue, after logging in to the VCSA6, I get the following error:
"Error occurred while processing request. Check vSphere WebClient logs for details."
I looked at the logs, and found out it's related to "Customer Experience Improvement Program", then a quick google search led me to the fresh KB2129053 - here: http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=2129053
I tried following the instructions, but I didn't do anything that could have triggered this except for following this blog post instructions.
William, do you have an idea how this can be resolved ?
Anyways, I updated my support case 🙂
I would recommend filing an SR for this if you haven't already
Hi William, restarted VSCA, and also filed an SR.
Weird thing is that the problem resolved on its own by some magic voodoo lol 🙂
Now I'm facing a Syslog alert about storage, but that's normal - an SR for that as well.
I can now say that we are fully upgraded to vSphere 6.0U1 and I didn't even plan it, just because I read your post.
How's that for adopting new technologies, aka, bleeding edge..
Running into this error in a test environment as well. I haven't even replaced the default certificates, so I'm wondering how to follow those instructions to fix it...
Ryan, I actually didn't do anything, it resolved on it's own.
As an added bonus, I enabled the CEIP.
Maybe try rebooting the VCSA again ?
Interesting. I left vCenter running overnight and when I checked this morning, it seemed to be able to connect to the CEIP (no more error when trying to load the status or when logging in, and properly loaded the CEIP status as "not joined"). However, when I rebooted vCenter, the CEIP and login errors come back. Are you able to restart without getting the error messages upon logging in? I'll check tonight and see if it clears up again...
Ryan, I didn't reboot VCSA after seeing VUM register correctly, but the CEIP issue for resolve on its own for some reason.
Thanks for the info. It seems like it does clear itself up after a few minutes/hours. I just checked again and the errors cleared up.
Just tried to update VCSA
Version 6.0.0 Build 2793784
Downloaded and connect ISO VMware-VCSA-all-6.0.0-3040890.iso
In ssh run command:
Command> software-packages install --iso
[2015-09-14T07:34:22.257] : Staging software update packages from ISO
[2015-09-14T07:34:22.257] : ISO unmounted successfully
[2015-09-14T07:34:22.257] : CD drives do not have valid patch iso.
[2015-09-14T07:34:22.257] : Installation process failed
What to do next?
Did you remember to connect the CD-Rom device from the VM settings ?
Yes 🙂
I understand my mistake.
I need PATCH iso.
Worked perfectly as described.
Many thanks.
I am unable to move vms from one server to another with vcsa 6.0u1.
When getting to the point to assign the network, I get stuck.
When reverting to 6.0 it works again.
Thanks again for this article. Since vSphere 6.0 port 5480 is not available anymore, I thought but I'm happy to see it's back in U1.
Thanks for this. Worked great. However, the VAMI is not running. Doing a "netstat | grep 5480" returns nothing. Is there a way to manually start it?
Did you reboot the system after applying the patch?
I did. Twice even.
I had the same issue. Check vami-lighttp service and it is not started. Once it is started, all works for me
This happens in case IPv6 is disabled on the vCSA. Edit /etc/applmgmt/appliance/lighttpd.conf accordingly. It should include server.use-ipv6="disable".
Regards,
Sebastian
Thanks!! That fixed it for me (now 5480 works)
Same problem with VAMI as well. Update finish successfully but VAMI is not working. I did notice that during Prossing software packages stage it show 142 packages. on the staged stage after that I do see only 134 packages. After the reboot I try to reapply the patch but it said that all of the packages are up to date and it wont continue. I did a little bit more digging and I found this error after running "/opt/vmware/share/vami/vami_config_net"
Traceback (most recent call last):
File "/opt/vmware/share/vami/vami_ovf_process", line 25, in
import libxml2
File "/usr/lib64/python2.6/site-packages/libxml2.py", line 1, in
ImportError: No module named libxml2mod
Traceback (most recent call last):
File "/opt/vmware/share/vami/vami_ovf_process", line 25, in
import libxml2
File "/usr/lib64/python2.6/site-packages/libxml2.py", line 1, in
ImportError: No module named libxml2mod
I have the correct version of the libxml2 module
mtlvc1:~ # rpm -q libxml2
libxml2-2.9.2-1.vmw.2748529
The problem maybe is mismatch between the Python libxml2.py and the VAMI libxml2 versions (2.7 vs 2.6).
Is Pyton supposed to be updated to version 2.7? Just guessing.
Hi Venti,
Ah, this is something another customer had reported. I believe it has to do with the patching taking a bit longer and didn't complete. Can you please file an SR, as I know GSS is currently involved in another case related to this topic. Thanks
Thanks William,
I didn't have any luck using the updated patch process to make VAMI working. I end up removing the current VC and deploy a new one from the main ISO image and everything is working fine.
I did the update on the external PSC using the update process first. It finished fine, but I'm curious now if it finished successfully based of this behavior on VCSA update. I know that you have listed as additional to VAMI in what is new in U1 that now we can access psc thru https:// VCSA-IP/psc. If do that I got a blank page (no errors). I can get to this options by using the https://FQDN_of_PSC/psc.
Is this is correct or is something wrong?
Another thing that I notice in your screenshot from VAMI you have your SSO Domain: vghetto.local and Status: RUNNING. In my case I have SSO Domain empty and Status: SSO is not initialized. I have 2 VC and they both show like that. Everything on PSC seems fine. https:///websso/ open the Welcome to the Platform Services Controller 1.0 page.
Any idea how to fix this SSO not running in VAMI? Do I need to deploy a new PSC from the original ISO to make sure that everything is on the same working level and repoint all of my VC's?
Thanks
Venti,
Depending on what you configured for the PNID (Primary Network ID) whether it's FQDN or IP, that'll be used within SSO when it does the re-directs to auth in. I would have thought IP should have automatically redirected to FQDN but honestly, I've always used FQDN when I connect. I'm curious if you're able to login to the vSphere Web Client? Perhaps SSO is still initializing ...
I have configured PNID to use FQDN. I don't have problem login with local administrator account either. Yes I can login to both VC's using the vSphere Web Client. for some reason SSO is not initialized. One more question. Under appliance settings there is a link to VMware Platform Services Appliance. If you hover over is going to https://psc-FQDN:5480. If I click on it it show a error that page cannot be loaded. Similar problem to VCSA VAMI problem after the update.
Is this link supposed to work?
Did you deploy VCSA w/External PSC?
Yes I did. I had external PSC and VCSA as version 6.0b before attempting the update.
Yea, I actually just deployed the same setup in my lab but didn't run into the problem you described. You're right, there was a first time login init that takes place when you login to the /psc endpoint, but once I got in, I was able to click into the VAMI and it took me to PSC VAMI w/o issues.
One thing I would like to point out, is that I had used IP vs FQDN. I don't know if that matters or not. If you're still having issues, I would say file an SR so GSS can take a look.
Thanks for you time to try to reproduce the error.
I decided to deploy a new PSC from the original ISO and it worked fine. I can get to VAMI without any problems and everything is working the way you described.
At least in my situation I can confirm that using the Patch ISO to update 6.0b appliances to 6.0 U1 it is not enables VAMI on VC or PSC.
I will redirect both of my VC's to the new PSC and then decommission the first PSC that is having these issues.
I have notice that when I login to PSC:5480 I do see now that SSO Domain is displayed and Status is RUNNING. When I go back to the VC that I already repointed to the same PSC is displaying the same like before that Status is SSO is not initialized.
On your screen shots above I do see that your setup is for Vcenter with embedded PSC.
Do you think that this is the reason that my SSO status is not initialized?
Can you confirm that with the Vcenter with external PSC the screen shot should look the same?
Thanks
Yes, I can confirm the status shows running for External PSC deployment which I also have in one of my environments and this is using FQDN as this environment has proper DNS. If you're still having issues, you can always file an SR and GSS can help take a closer look.
I have this same problem. Upgrade from 6.0b seems to complete fine. Upon reboot, VAMI is not accessible. libxml2 version not 2748529, but is stuck at 2434084. I've reverted to pre-upgrade snapshot twice and get the same result every time. It also appears that others are having this issue: http://www.settlersoman.com/problem-with-vcenter-vami-interface-after-upgrade-to-vcsa-6-0-u1/
This is just one reason why it would be desireable to still have the option of updating via the old method, via a fresh appliance and migrating data from 6.0 to 6.0 U1.
My upgrade didn't go so well. Upon upgrading the appliances (external psc) and rebooting (first PSC, then vCenter) I was unable to login to the Web Client with my AD account. When I logged in as *protected email* I discovered that my PSC was no longer a member of the AD Domain. Here's how it happened...
Checking the identity source in SSO admin I get the error: The name of the identity source does not match the existing Integrated Windows Authentication identity source.
When trying to re-add the domain as an SSO identity source I got the error: The vCenter Single Sign-On server is not currently joined to any domain. You cannot complete the current operation.
When I try to rejoin the domain I get the error - Idm client exception: Error trying to join AD, error code [67].
I can't find that error code ANYWHERE.
Any ideas?
Thanks,
Joe
Hi Joe,
Sorry to hear the upgrade didn't go well. If you've not already, I would recommend filing an SR, GSS can definitely help take a look to see what went wrong.
I cannot seem to change the URL below, does not allow me to type anything in the box
I also tried "Use specified repository " which accepts input but gives me errors. Any ideas?
Repository URL
https://vapp-updates.vmware.com/vai-catalog/valm/vmw/647ee3fc-e6c6-4b06-9dc2-f295d12d135c/6.0.0.5120.latest/
Thanks for the article, I used this to update a 6.0GA appliance to 6.0U1. One note for the n00bs would be that to wait maybe another 5 minutes after you can access the VAMI before attempting to logon. Otherwise you may see a lot of odd unresolved references. Also, the logon is root and not integrated with the SSO auth.
I too am getting "SSO is not initialized" in the status section of the summary section of VCSA management UI. This was a clean install as I am beginning to do some testing with the vCenter 6 U1 appliance.
vCenter seems to be functioning properly with no errors under the services section. Any suggestions?
I had open a ticket to VMware and he confirm that to access PSC https:// FQDN of VCSA or IP/psc is the correct url if we have an embedded install of PSC and vCenter, if PSC is external you need to use https://FQDN of PSC or IP/psc.
For SSO Domain comming up as empty and Status reporting SSO is not initialized is interesting,can we go on webex and look into the issue.
We had webex and he saw what I have in place. He decided to replicate our environment but was not able to complete the reproduction of the issue.
I decide then to rebuild everything again from the begging. I have 1 PSC and 2 VC's deployed from the original Vcenter 6 U1 ISO and vCenter seems to working fine. I can access everything thru SSO or VAMI, but in summary section of VCSA management UI I still have “SSO is not initialized”.
According to William he sees the status running in his environment, but I don't in mine. I don't know is this is true statement now. I guess I will reopen the case and try to see if someone can find the problem.
Yes, this is exactly my problem. The SSO Domain field is blank and status shows as "SSO is not initialized". Maybe it is a bug?
I did open another case to VMware and SC confirm that in External PSC deployment "SSO is not initialized" is how supposed to be displayed in any VC. The explanation of that is that SSO service is not running on this VC because is handle by the external PSC. It will be initialized only in Embedded deployment. So I think your and mine deployments are fine.
We're also seeing this error message after upgrading. Everything works fine though. After opening a case with VMware they're telling us to repoint vcenter to SSO (kb 2113917).
Proxy setting in VAMI GUI interface seems not to work correctly. Only the HTTP_PROXY variable in /etc/sysconfig/proxy is added not HTTPS. But the repository is https://....
Yea, it looks like only HTTP proxy has been implemented in the VAMI UI. I've already let Engineering know about this. In the mean time, if you wish to use HTTPs proxy, you can do so by using the appliancesh command to either HTTP, HTTPS & even FTP proxy.
appliancesh supports multiple http/https/ftp proxy.
For details explore appliancesh help for networking.proxy
Hi William, thanks a lot for the instructions. Worked like a charm 🙂
Even though I like to use the CLI I'm glad to see the VAMI WebGUI is back.
However it seems to have a localization issue with the update settings. It only works when I set my browser to english.
The GUI works fine in german and (most of) the content is properly translated, including the week days in the update settings. But if it's "jeden Tag" or "Montag" instead of "Everyday" or "monday" it reports back an error because it expects the english words for the settings...
(Failed to set update repo info: Invalid day. Should be a day of the week or Everyday.)
So in case anybody else encounters this issue, for now you just have to set your browsers preferred language/locale to English before accessing VAMI WebGUI.
Regards, CM
Tried to change update URL and recieved an error "Failed to set update repo info: Internal Error".
Setting https proxy "com.vmware.appliance.version1.networking.proxy.set --protocol https --port proxyport --server proxyservername" resolved this problem.
I'm having the same issue, where did you update this setting?
In VCSA shell
Found thx
Thank you for this post. The upgrade of my lab was smooth.
But when I go to the Web Client, the Adobe Flash clock animation spins for a little while, then it gives the error "An internal error has occurred - Index '0' specified is out of bounds." Happens in IE10 and Chrome.
It is a standalone PSC and I can login to the https://:5480 appliance management and it looks healthy. I can deploy a new VCSA from scratch, but it would be nice if I could fix it.
That sounds strange as I have this exact setup in one of my lab setup. I would recommend filing an SR w/VMware GSS to see why you might be having this problem
Shutdown command complained about syntax. I had to run following:
shutdown reboot -r "Updated to vSphere 6.0u1"
There are lot of article that talks about VMware-vCenter-Server-Appliance-6.0.0.10000-3018521-patch-FP.iso for updating existing VCSA 6 to 6. u1 but I don't see that download in vmware site. I only see full iso VMware-VCSA-all-6.0.0-3040890.iso. Does the process of upgrading with Software-package install would work with VMware-VCSA-all-6.0.0-3040890.iso?
Did you have a careful look at the article above? It tells you exactly where to download the the patches 🙂
Few things:
a) I was not able to get to https://VC:5480 after patching PSC and VC (external PCS running as appliance, VC running as appliance) to update 1.
I run netstat -tnl | grep 5480 and nothing was listen on tcp 5480.
Applmgmt service was running and restarting it did not fix issue
Fix: Reboot VC and make sure IP6 is enabled.
b) In Appliance management (https://VC:5480) SSO domain is blank and status is : SSO is not initialized.
Everything works fine. I run following to check PCS , site and domain on VC - no issues found
/usr/lib/vmware-vmafd/bin/vmafd-cli get-ls-location --server-name localhost
/usr/lib/vmware-vmafd/bin/vmafd-cli get-site-name --server-name localhost
/usr/lib/vmware-vmafd/bin/vmafd-cli get-domain-name --server-name localhost
I will open SR with GSS.
Hi William,
In a multi site environment using stretched SSO with external PSC and VC at each site is it safe to assume that I upgrade all the PSCs first then move onto the VCs?
Cheers,
Gerg
Yes, or you can just update a single site (since PSC are backwards compat) and get both PSC/VC updated before moving on. Either options would work, as long as you update PSC first before upgrading VC
Thanks for the speedy response:-)
One site at a time it is then.
Cheers,
Gerg
Actual update repo link:
https://vapp-updates.vmware.com/vai-catalog/valm/vmw/dcce55fe-2402-11e2-af10-000c29b4e3e7/6.1.0.0.latest
Hi William
I need to upgrade my environment to 6.0U1. I currently have 5.5.0.30100 Build 3154314 (I think it is a 5.5U3, it was originally a 5.5 which I have updated to 5.5U3 using 5.5 VAMI UI).
I was honestly thinking to deploy from scratch the 6.0, configure it and then upgrade to 6.0U1. My environment is not so huge (19 hosts) so it would not be a big deal to add again the ESXi hosts to the new 6.0.
Having said that, I have seen around several guides on how to upgrade from 5.5 to 6.0 and I was curious to give a try if you recommend it. Can you tell me if the steps below are correct and eventually suggested? 🙂
1) run the upgrade as instructed on the VMware deployment guide on page 25
https://www.vmware.com/files/pdf/techpaper/vmware-vcenter-server6-deployment-guide.pdf
2) provide the new vCenter name and the details of the old one
3) provide temporary ip address for the new vCenter which will get the old ip from the previous one when all the data has been transferred and the old one will be shutdown
4) when the new one is up and running it can be updated to 6.0U1 using the guide on this page
Thanks
Mario
Hi William, I have been some troubles upgrading my VCSA 5.1u3 to VCSA 6 u1, when I checked the upgrade logs seems to be database postgrees error
Anyway, I want to try a different way. Is it recommended remove my ESXi(5.1) hosts from the actually vcenter and re-add these hosts in a fresh install of VCSA 6.0?
Can I do it without VM downtime ?
Does it possible ? Or I will have problems with register vpxd, VPXA hostd?
Thanks
Do I have to change the repo link before I update to 6.0U1b? I see that this version was released today and I can download the ISO, but the appliance does not find any new versions with the check URL button.
If you've upgraded previously and did not get the version with updated URL, then you'll need to change it else you won't see newer updates
I changed the URL to the one you mention in your article before. But it just took some time until the repo was update by VMware, now I see the update for 6.0U1b.
Hi,
I'm failing to update VCSA 6.
I'm trying to update vcsa 6, but it won't.
screen output:
Command> software-packages stage --iso
[2016-02-14T20:46:47.045] : Packages already staged.
[2016-02-14T20:46:47.045] : Run software-packages list OR software-packages install to continue
[2016-02-14T20:46:47.045] : Run software-packages unstage to delete old staged packages
Command> software-packages unstage
[2016-02-14T20:46:57.045] : Unstage operation completed successfully
Command> software-packages stage --iso
[2016-02-14T20:47:01.045] : Staging software update packages from ISO
[2016-02-14T20:47:01.045] : ISO mounted successfully
[2016-02-14 20:47:01,364] : Running pre-stage script.....
[2016-02-14T20:47:02.045] : Verifying staging area
[2016-02-14T20:47:02.045] : Validating software update payload
[2016-02-14T20:47:02.045] : Validation successful
Command> software-packages list --staged
[2016-02-14T20:48:39.045] :
category: security
kb: http://kb.vmware.com/kb/2111640
vendor: VMware, Inc.
name: VC-6.0.0a-Appliance-FP
tags: [u'']
productname: VMware-vCenter-Server-Appliance
releasedate: April 9, 2015
version: 6.0.0.5110
buildnumber: 2656759
rebootrequired: True
summary: Patch for VMware vCenter Server Appliance 6.0
severity: critical
Command> software-packages install --staged
[2016-02-14 20:49:06,957] : Running test transaction ....
[2016-02-14T20:49:07.045] : Installation process failed
anyone any idea?
Hello Wiliam
I got the same problem when updating VCSA and PSC U1 to U2 using FP iso. Update failed and was frozen using both cli and vami options. I had IPv6 disabled in the VMs and I enabled IPv6, restarted VMs and reattempted the upgrade.
The upgrade completed successfully.
Command> software-packages install --iso --acceptEulas
[2016-05-25T16:03:53.146] : Staging software update packages from ISO
[2016-05-25T16:03:53.146] : CD drives do not have valid patch iso.
My current version info
Type:
vCenter Server with an embedded Platform Services Controller
Product:
VMware vCenter Server Appliance
Version:
6.0.0.10000
I mounted VMware-vCenter-Server-Appliance-6.0.0.20000-3634791-patch-FP.iso
Im trying to run the update from 6.0 GA to 6.0 U2 and its failing saying that the staging area does not have sufficient space. Trying to figure out how to increase the staging area space. Any idea how to check or expand this?
I'm having the same issue. Anyone found a solution to this? There's virtually no documentation of this on vmware.com or the web.