ESX 3.5U4 Upgrade and Issues Thread

[ufo8mydog]

Hi all,

Now that U4 has come out, has anyone had a chance to be a guinea pig and test it - have you had any issues, bugs, or weird stuff happen? (well, a lot of weird stuff happens with ESX. It is just a matter of degree ).

[Zahni]

Hm,

may try his:

(for BL 465c G% in a C7000)

• Make sure, the OBA has the firmware 2.41

• Make sure, you had ILO2 Firmware 1.70 installed:

• copy hpmgmt-8.2.0-vmware3x.tgz to /tmp/hp

• untar it

• got hpmgmt/820

• run ./installvm820.sh --uninstall

• after that, remediate the Host

• After the final reboot:

• Download the "HP BladeSystem Firmware Deployment Tool" version 1.50

• Mount it in ILO's remote console and boot from it

• After that, run ./PreInstall_Setup.sh and reboot (!!!)

• finaly, run ./installvm820.sh --install

This had worked for me.

[mrray]

I have an issue..

I use NFS for storage connected through redundant NetXen 10Gb cards. (in this case they areidentified as HP NC510F PCIe 10 Gigabit Server Adapter

After the upgrade I can't get IP traffic on these (worked fine with the esparate driver in Update 3)

The cards are there, I have link at 10.000 Mbps and there is even a few packets in tx/rx:

(both these cards are linked to a VMkernel vSwitch and connected to a layer 2 vlan (no routing)

vmnic10 Link encap:Ethernet HWaddr 00:17:A4:3E:0C:E0

UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1

RX packets:4498 errors:0 dropped:0 overruns:0 frame:0

TX packets:0 errors:0 dropped:0 overruns:0 carrier:0

collisions:0 txqueuelen:1000

RX bytes:323543 (315.9 Kb) TX bytes:0 (0.0 b)

Interrupt:209

vmnic11 Link encap:Ethernet HWaddr 00:17:A4:3E:1C:28

UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1

RX packets:4513 errors:0 dropped:0 overruns:0 frame:0

TX packets:1054 errors:0 dropped:0 overruns:0 carrier:0

collisions:0 txqueuelen:1000

RX bytes:324958 (317.3 Kb) TX bytes:67456 (65.8 Kb)

Interrupt:201

The Storage system is a NetApp.

The servers are HP D580 G5s...

What I tried:

Install the driver that I originally installed (http://vmware.com/support/vi3/doc/drivercd/esx350-unm_nic-4.0.230-146051.html)

Try a fresh install of 3.5 U4 on a host.

Anyone come across this before? Any ideas?

[kinny_k]

Ah, thanks for pointing that out. I see in /etc/xinetd.d/vmware-authd that authd is supposed to be running on 902, and indeed is according to netstat.

I didn't change anything myself. I'm not sure where this change happened, but I'm sure I didn't do it intentionally. Could there be some bug in the update process?

Looking at /etc/vmware/config I see that authd.client.port is still set to 902.

With things set the way they are I don't see how changing the config will help, so I will first try to reinstall the VMware-vpxa as you recommend. If that doesn't work I'll go backwards and change things to 903, since that is where things seem to want to connect.

I think the right solution would be finding where vmware-authd is configured to redirect the mks to port 903, and fixing that. That should be what I try first. anyone know where that setting is?

Thanks.

[hollandwl]

All our NFS datastores became inactive and every setting that we had changed on the host went back to defaults. This should NEVER happen during an update/upgrade.

NFS.MaxVolumes was reset from 32 to 8

Net.TcpIpHeapSize was reset from 30 to 9

Net.TcpIpHeapMax was reset from 120 to 30

Setting these back from their defaults to our settings and rebooting did not fix the issue with the datastores being inactive. The funny thing is, you can browse the datastores through the VI Client.

One of my colleagues has opened a ticket with VMware, we'll see if they come up with anything.

[RadioMarky]

Hey all, I've just downloaded/installed a fresh install of the latest 3.5u4 and it's running fine on a HP Proliant ML350 G5. Will post if anything strange happens in the near future.

Am about to update our main ESX server tonight to u4. I can't see a lot (if any) major changes to our setup from what i've read in what the update actually contains etc. Again, will let you guys n gals know if any horror stories.

RM

[Blizz]

I was testing ESX 3.5u3 on a Dell PowerEdge R00 (ICH9 Controller). It worked flawlessly. I have a 150GB VMFS partition on HDA (HDA3) and a 500GB VMFS partition on HDC (HDC1).

Today I upgraded to U4 and upon loggin in on Infrastructure I got the "... does not have persistent storage" dialog box. No VMFS partitions are visible, they are still there.

Installed the ESX again, updated to U3, everything worked perfectly. Ran de U4 again,again the no storage thing.

Under storage adapters, only the iSCSI is visible in U4, whislt U3 has 2 others.

The R200 only has ATA mode or Off for the SATA devices.

Any clues on how to fix this?

[kinny_k]

Well I tinkered a bit an now it works. I'm not sure what exactly did it, but I suspect it was your suggestion to add 'vmauthd.server.alwaysproxy = "true"' to /etc/vmware/config.

I tried everything else you suggested. Even going as far as to remove and re-add the host. After changing the port back to 902 it's working.

I've run into something else now, bit odd. I've lost the information for that host. General, Resources, and the tasks, those panes aren't complete. General has no information filled in, Resources doesn't show the Datastore's, but it does show the CPU and Memory usage, and the tasks pane shows everything.

This isn't a serious problem, but I find it annoying. Any idea on what I've done now?

Thanks.

[eagleh]

hi bitsorbytes, how did you manage to disable the agents if you couldn't get the server boot up? thanks.

[eagleh]

Anybody can confirm if the HP Agents 8.20 will cause ESX host restart automatically? 8.1.1 did and gave me a heartattack!!! Thanks.

[deceit]

Hi eagleh,

8.20 install hasnt rebooted any of my BL465cs.

If you need to cancel the start up process because it gets hung. Wait for the "Press I for Interactive" prompt to be displayed. Press Y for all services but HP and you should be right.

[bitsorbytes]

hi bitsorbytes, how did you manage to disable the agents if you couldn't get the server boot up? thanks.

During the booting of ESX, press 'I' key when prompted to do a interactive boot. It will ask you to confirm loading of each service. Confirm yes to every service other than the hpxxx agents and the server will boot fine.

From here, do a uninstall of the agent from the commandline or update to 8.20 of the agents.

[eagleh]

Thank you deceit.

[eagleh]

Thank you bitsobytes!

During the booting of ESX, press 'I' key when prompted to do a interactive boot. It will ask you to confirm loading of each service. Confirm yes to every service other than the hpxxx agents and the server will boot fine.

From here, do a uninstall of the agent from the commandline or update to 8.20 of the agents.

[timparkinsonShe]

eagleh, 8.20 has restarted the one ESX server (DL585 G2) I've put it on so far. Oddly about 5-10 minutes after the install is succesful and returns a prompt. It also appears as a system halt going by the message on my SSH client. All a little worrying really, I'm glad I hadn't brought it out of maintenance mode!

[depping]

I haven't personally experienced any issues yet...

Duncan

VMware Communities User Moderator

-

---

Blogging:

Twitter:

If you find this information useful, please award points for "correct" or "helpful".

[DGI_Drift]

Hi

Updated to ESX 3.5 Update 4 at one host. Can't see that I have any problems yet.

But I update HP-SIM agent from 8.1.1 to 8.20 aswell, but now I can't get any info from the management homepage.

The homepage default page is in the picture.

Some other that have this problem, and have a solution?

I uninstalled first, the installed. I have done it 2 times, but still nothing.

[timparkinsonShe]

DGI_Drift, I had the same problem. The mistake I made was I told it I already had an snmpd.conf not realising that the uninstaller for the previous version had reverted it. I just copied one from another host and restarted snmpd.

[eagleh]

timparkinson, Thanks for the warning. actually that was what happened to me when i installed 8.1.1 (DL360 G5). To be safe, as you mentioned, hold it in maintenance mode for a while after 8.2.0 has been installed then.

eagleh, 8.20 has restarted the one ESX server (DL585 G2) I've put it on so far. Oddly about 5-10 minutes after the install is succesful and returns a prompt. It also appears as a system halt going by the message on my SSH client. All a little worrying really, I'm glad I hadn't brought it out of maintenance mode!

[eagleh]

I suggest playing with smpd.conf file or services.

[MBreidenbach0]

I've got a HP ML110 G5 with SATA drives as VMware ESX test server toy (it's slow but works). I believe it has ICH9 too. After updating from ESX 3.5 U3 to U4 it didn't see any VMFS volumes.

Then I enabled SATA RAID in BIOS (which didn't work with U3) the onboard SATA RAID controller told me it sees two legacy SATA drives and after ESX hat booted my VMFS volumes were back.