VMware Cloud Community
TB2010
Contributor
Contributor

Manual updating using VMRoyal directions help

I am following the directions located here: http://communities.vmware.com/blogs/vmroyale/2009/10/12/single-use-esxupdate-how-to-for-esx-4

I place the patch in the datastore for the host.

I get to the point were I run this command: mv /vmfs/volumes/datastore/ESX400-200909001.zip /var/updates

I get an error: mv: cannot stat '/vmfs/volumes/datastore/ESX400-200909001 no such file or directory

I can not cd to the datastore directory on the server or browse it. I know I am missing something and my knowledge of Unix is limited. Smiley Happy

when I do a vdf -h I do see the volumns and drive size. I do see the host volumn as:

/vmfs/volumns/48ec7622-..... then size 26g size 8g used /vmfs/volumns/vmhostname:storage1

Any help would be appricated Smiley Happy

0 Kudos
8 Replies
vmroyale
Immortal
Immortal

Hello.

I get to the point were I run this command: mv /vmfs/volumes/datastore/ESX400-200909001.zip /var/updates

I get an error: mv: cannot stat '/vmfs/volumes/datastore/ESX400-200909001 no such file or directory

You will need to use your own unique datastore name, if you haven't done so.

I can not cd to the datastore directory on the server or browse it. I know I am missing something and my knowledge of Unix is limited. Smiley Happy

Are you logged in as root, or did you "su -" after logging in?

What happens if you just issue a "cd /" and then a "ls" ?

Good Luck!

Brian Atkinson | vExpert | VMTN Moderator | Author of "VCP5-DCV VMware Certified Professional-Data Center Virtualization on vSphere 5.5 Study Guide: VCP-550" | @vmroyale | http://vmroyale.com
0 Kudos
TB2010
Contributor
Contributor

I am logged in as root, I can ls and see what is in the directories and cd. I can not cd to any of the datastores. I use the command:

cd /vmfs/volumes/serverstore (serverstore is the datastore name)

I get the error

-bash: cd/vmfs/volumes/serverstore no such file or directory

when I do the vdf -h I see all the directories including the datastores but cannot connect. I am wondering maybe I need to mount the datastore since it is on a central filestorage unit? I have also tried putting it on the host datastore which is in my first post with the same error.

0 Kudos
khughes
Virtuoso
Virtuoso

The process I used to manually update ESX hosts before we moved to Update Manager (which you should look into if you can), is I used a program like WinSCP to upload the files to the /tmp folder on the ESX host, from there was able to "cd /tmp" and then "ls" and it would list all the updates that were there. From that point you should be able to follow the rest fo the procedure on unpacking and then installing each of the updates.

  • Kyle

-- Kyle "RParker wrote: I guess I was wrong, everything CAN be virtualized "
0 Kudos
vmroyale
Immortal
Immortal

Can you post the output of the "vdf -h" command?

Brian Atkinson | vExpert | VMTN Moderator | Author of "VCP5-DCV VMware Certified Professional-Data Center Virtualization on vSphere 5.5 Study Guide: VCP-550" | @vmroyale | http://vmroyale.com
0 Kudos
TB2010
Contributor
Contributor

I can not post the out put. But I can recreate some it so you get an idea.

#vdf -h

Filessystem Size Used Avail Use% Mounted on

/dev/sda5 5G 1.7G 3.2G 37% /

/dev/sda4 2G 70m 1.8G 4% /var/log

... (more entries)

...

...

vmfs/devices 40G 0 40G 0% /vmfs/devices

/vmfs/volumes/40sec785-655464457-e2e4-002a498324

25G 8G 17G 30% /vmfs/volumes/vmhost1:storage1 --- this is the actually host server

/vmfs/volumes/40sec490-695267567-e2e4-03a9032343

20G 5G 15G 20% /vmfs/volumes/serverstore -


datastore were I placed the file

I have also tried placing the file on the vmhost1 itself and get the same error:

cannot stat '/vmfs/volumes/vmhost1:storage1 no file or directory

or

cannot stat '/vmfs/volumes/serverstore no file or directory

0 Kudos
khughes
Virtuoso
Virtuoso

Shesh Brian and they made you a vExpert with that procedure on how to manually update the ESX hosts...

Here's a procedure I wrote for my company when we were on ESX 3.5, but it should essentially be the same for ESX 4.0:

Procedure

Downloading the Updates

1. Review updates from VMware’s website:

(http://support.vmware.com/selfsupport/download/)

2. Download patches that will need to be installed onto the ESX hosts and save them to your local hard drive.

3. Use WinSCP to transfer the downloaded updates to the ESX host you are going to update.

(http://www.winscp.net)

Upload all the update files in /tmp folder. Once the transfer is complete you can close down WinSCP.

Preparing the ESX Host

4. With the Virtual Infrastructure Client, under the hosts and clusters view, select the target ESX host. On the Summary tab, find the Commands box and click ‘Enter Maintenance Mode’.

Uncompressing and Installing the patch

7. Open up a putty session with the ESX host you are going to update, logging in with the root account.

(http://www.chiark.greenend.org.uk/~sgtatham/putty/)

Change directories to where your updates are located, /var/updates by using the command:

  1. cd /tmp

NOTE: Once in the /tmp location you can verify that the files are there by issuing the command:

  1. ls –l

8. Uncompress the patch files by running the command:

#unzip esx350-.zip

NOTE: This will create a folder within the /var/updates folder with the name of the patch.

9. Change directories to the new folder containing the patch:

  1. cd ESX350-<patch number>

If you are going to be doing multiple update installs you will want to use the switch to force it not to reboot after installing each patch (if the patch requires it).

10. To install the patch and force it not to reboot after installation use the command:

  1. esxupdate --noreboot update

NOTE: If this is going to be your only update or the last update in the multiple patches you can use the same command without the noreboot switch in it:

  1. esxupdate update

NOTE: It is best practice to reboot the ESX host after installing new patches (even if it may not require them). If the patches do not require a reboot, you can force a reboot by using the following command:

reboot

11. Repeat steps 9 – 11 for any remaining updates that need to be installed.

12. Once the host has been fully patched and rebooted, take the ESX host out of maintenance mode and use DRS to determine what servers should be moved back to the ESX host.

*END*

Obviously this was customized for a monkey to do it at our office if I wasn't around, but you can get the main swing of things.

  • Kyle

-- Kyle "RParker wrote: I guess I was wrong, everything CAN be virtualized "
0 Kudos
vmroyale
Immortal
Immortal

Possibly stating the obvious here, but make sure there are no capitalization differences or spaces in any of your file or directory names. If nothing else, I will have to agree with Kyle about just using WinSCP to move the files around. It might be faster than figuring out what is actually going on here.

Brian Atkinson | vExpert | VMTN Moderator | Author of "VCP5-DCV VMware Certified Professional-Data Center Virtualization on vSphere 5.5 Study Guide: VCP-550" | @vmroyale | http://vmroyale.com
0 Kudos
TB2010
Contributor
Contributor

Still working on problem. Got little side tracked fixing other things. Going to redo everything from step one and also verify sentences.

Not sure if we are allowed to use winSCP so trying to get it to work first without going that route.

Will update with results.

0 Kudos