AntonVZhbankov
Immortal
Immortal

vSphere 4u1 - any issues?

Please post any issues with 4u1 here


---

MCSA, MCTS, VCP, VMware vExpert '2009

http://blog.vadmin.ru

EMCCAe, MCITP: SA+VA, VCP 3/4/5, VMware vExpert http://blog.vadmin.ru
0 Kudos
70 Replies
KarlKatzke
Contributor
Contributor

After downloading the VMWare-VIMSetup-all-4.0.0-208156 zip file to a computer with no previous installation of vSphere, running the installer, and connecting to the server that I'd already upgradeed, my vSphere client reports that client and server are still build 208111 in help/about...

0 Kudos
DSeaman
Enthusiast
Enthusiast

<Deleted>

Derek Seaman
0 Kudos
Oletho
Hot Shot
Hot Shot

PSOD on boot after updating with Update Manager

Ole Thomsen

0 Kudos
f_pantazis
Contributor
Contributor

i have a issue with the datastore browser. Just upgraded from vcenter 4.0 and esx 4.0 to U1 and the datastore browser does not display folder or files

(from within the virtualcenter)

directly connecting to the esx server and browsing the datastore works fine.

tried rescanning/rebooting and still nothing...

anyone having the same problem?

0 Kudos
Jain
Hot Shot
Hot Shot

this bug should have fixed.

--------------------------------- *Jain* | VCP4 | +Please award points, if you found answer useful+
0 Kudos
DSeaman
Enthusiast
Enthusiast

Updating the VUM SSL certificates is still broke with U1, but I found a way around the problem. Basically you de-install VUM and re-install it with your custom SSL certificates and it appears to work.

Full details are at:

http://derek858.blogspot.com/2009/11/vcenter-update-manager-40-ssl.html

VMware fixed the thumbprint problem with vCenter in U1, but didn't bother to fix VUM.

Derek Seaman
0 Kudos
GeneNZ
Enthusiast
Enthusiast

Followed your guide to the letter RE the VUM Certificates, but no luck. Tried it twice, with a reboot included and again it appears to retain the VMWare Self-signed certificate.

Difference is I'm using Win 2008 32bit, and I'm using a CA signed certificate. Still getting the following in the VUM logs:

SSLStreamImpl::BIORead (75ACFB0) failed: The specified network name is no longer available.

SSLStreamImpl::DoServerHandshake (75ACFB0) SSL_accept failed with BIO Error

Vmacore::Http::HttpSvcImpl::AcceptHttpConn accept failure - no read scheduled: ,SSL Exception: BIO Errorclass Vmacore::Exception *

Vmacore::Http::HttpSvcImpl::AcceptHttpConn stream is NULL - no read scheduled

Problem I'm seeing is that I can't appear to create a Upgrade Baseline from my vCenter (which has been upgraded) after I upload the ESX4U1 iso and zip. I'm receiving the following error in the VUM logs:

Method integrity.UpgradeProductManager.importProduct threw undeclared fault of type integrity.fault.HostUpgradeChecksumFailure

On the Upgrade Baseline interface it just says: "File Upload Error". I have checked the checksum of the downloaded file (using MD5sum) and it is indeed correct. I was wondering if it had anything to do with my certificates?

0 Kudos
DSeaman
Enthusiast
Enthusiast

Interesting, I'm not sure what to say. My base system was server 2003 R2 x64 and it was dedicated to VUM. Like my blog said, I just did a typical VUM install, swapped the certs, de-installed VUM and re-installed. Did you make custom SSL certs for vCenter and did they work?

Derek Seaman
0 Kudos
GeneNZ
Enthusiast
Enthusiast

My VUM is part of my vCenter (thats another difference). We created a certificate, but it is not self-signed, it is CA-signed (by RapidSSL). Worked perfectly fine for vCenter, but not for VUM.

Wondering if I have to adjust anything using ADSI Edit in ADAM etc? Your blog post mentioned that the install should adjust the ADAM thumbprint, so I'm at a loss? Either way, I'm not even sure if the certificate has anything to do my upgrade baseline, but now I've got two annoying issues.

Any additional help or thoughts will be be great!

0 Kudos
DSeaman
Enthusiast
Enthusiast

You can go into the ADAM database and check the thumbprint for the VUM service. It's pretty hard to find with LDP, but it's burried in there. For my vCenter installation the VUM thumbprint is located at the following DN in the ADAM database. The GUID will be different for you, but the rest of the path should be the same.

CN=com.vmware.vcIntegrity,CN=35143651-209D-4D5E-873C-20617A09B3DE,OU=ComponentSpecs,OU=Health,DC=virtualcenter,DC=vmware,DC=int

Derek Seaman
0 Kudos
GeneNZ
Enthusiast
Enthusiast

Nope this one has me stumped. I went into ADSI Edit and found the thumbprint and deleted it, rebooted and same problem. I went through this procedure because it was the same process I went through to fix my vCenter Service Status (as shown in this thread: http://communities.vmware.com/thread/226681). It was very much a sledgehammer approach, but worth a try. Still haven't found a fix with my upgrade baseline issue either. Will open a support case with VMware.

One thing I have noted is that all my alarms were duplicated. I had to go through the alarms section and delete the duplicates.

0 Kudos
enDemand
Enthusiast
Enthusiast

Hope this is a rare issue, but for me, this one was quite impactful. After the complete upgrade (ESX 4.0 to U1 and vCenter Server 4 to U1), I realized that all of my resource pools and the few vApps that I had were gone. The VMs were still intact, but configured in a flat resource hierarchy. I initially noticed when jobs targeting those missing objects began failing. I'm assuming that the issue was a result of the vCenter Server upgrade and not the ESX upgrade.

Anyone else seen anything similar to this?

David

If you find this or any other answer useful, please consider awarding points by marking the answer "correct" or "helpful".
0 Kudos
krowczynski
Virtuoso
Virtuoso

I've updated both of my hosts (Dell R410) with VUM with no problems.

Everything went fine!






MCP, VCP

MCP, VCP3 , VCP4
0 Kudos
MaciejKrol
Contributor
Contributor

Hi,

we have upgraded few days ago from ESX3.5 U3 to ESX4.0 U1 without any issues - everything works fine. It was on Intel Platform SR2500 with local storage on Adaptec 5805 (firmware 16501) - we used hostupdate utility.

0 Kudos
HenrikElm
Contributor
Contributor

My upgrade from 3.5 U4 to 4.0 went just fine. Did it at customers too with no problem. 4.0 to U1 with VUM went just fine too.

/Henrik

0 Kudos
Jean_Paul
Contributor
Contributor

Just updated vSphere server 4 to Update1 and updated Update manager & Converter. Works as it was expected.

Created a new upgrade baseline and tried to update an ESX4.0.0. 175625 .... Bummer!

"Upgrade is not supported from host version 175625 to release version 4.0.0 Update 1 Build 208167"

Why??? It looks like a normal 'upgrade' to me.

0 Kudos
chrisaug
Contributor
Contributor

I got the same message. I found KB article 1013068 on kb.vmware.com that said if you're not running any Cisco Nexus this conflict warning can be safely ignored. I continued with the update and everything went fine. Here's the link to check out for yourself. Good luck

http://kb.vmware.com/selfservice/microsites/search.do?language=en_US&cmd=displayKC&externalId=101306...

0 Kudos
Jean_Paul
Contributor
Contributor

Thanks. Hmm, sounds like something to thest in the weekend.

We don't use the Nexus1000.

It's just an ESX4-drs cluster (3 hosts/ 500VM's -> 200+ running) with the latest patches since this weekend.

0 Kudos
chrisaug
Contributor
Contributor

Yeah, we are a similar env. 3 hosts, Ent Plus with HA, DRS, dvSwitches, & ~50 VMs connected to NetApp SAN using NFS. I tested it with the first host and all went well. I actually had other updates too so I installed all updates but the two in conflict, rebooted the host, etc... Then I did the two updates in conflict and they both installed without any issue. I followed this procedure for two of my three hosts. The third I did all the updates at once and that worked as well.

0 Kudos
enDemand
Enthusiast
Enthusiast

Glad to hear other upgrades went well. Doesn't seem like it's a problem inherent to all upgrades, but there's no doubt that I had to recreate ~10 resource pools and 3 vApps. Something happened. The upgrade via VUM itself went smooth with no issues. Afterwards, I did the VC4 upgrade, including Orchestrator...didn't botch up anything that Orchestrator had...all went smooth. It wasn't until I connected from my workstation and upgraded the client that i noticed that the RPs and vApps were gone.

If you find this or any other answer useful, please consider awarding points by marking the answer "correct" or "helpful".
0 Kudos