Difference between revisions of "Update vzkernel newbie"

From vpsget wiki
Jump to: navigation, search
(Created page with "Update only vzkernel on Centos. Check the bootlist: cat /boot/grub/grub.conf Proceed with update: yum update vzkernel Check the bootlis again and note that new kernel appe...")
 
 
(6 intermediate revisions by the same user not shown)
Line 1: Line 1:
Update only vzkernel on Centos.
+
*UPD this also may be applied to openvz7 / virtuozzo7
 +
*ADD consider to complete yum update
 +
  {yeahh the full update} even if you updated only kernel and noted weird errors while starting the CTs or/adn VMs - try to complete full yum update after and restart vz/qemu service even w/o reboot *(as reboot required only for kernel update apply)
 +
 
 +
Update only vzkernel on Centos. Read till the end before proceed.
  
 
Check the bootlist:
 
Check the bootlist:
 
  cat  /boot/grub/grub.conf
 
  cat  /boot/grub/grub.conf
 +
if you using newer grub:
 +
cat /boot/grub2/grub.cfg
 
Proceed with update:
 
Proceed with update:
 
  yum update vzkernel
 
  yum update vzkernel
Line 11: Line 17:
 
  vzlist | wc -l
 
  vzlist | wc -l
 
You may save the output in order to be able to compare it with the same command output after reboot/all vm start
 
You may save the output in order to be able to compare it with the same command output after reboot/all vm start
  vzlist | wc -l >/tmp/vzlist.txt
+
  vzlist | wc -l >/home/vzlist.txt
  vzlist >>/tmp/vzlist.txt  
+
  vzlist >>/home/vzlist.txt
 +
 
 +
for virtuozzo use:
 +
prlctl list >/home/prlctl.txt
 +
to check/save only openvz7/virtuozzo7 VMs
 +
prlctl list --vmtype vm | wc -l >/home/prlctl_listVMS_wc-l.txt
  
 
Reboot:
 
Reboot:
Line 19: Line 30:
 
Once reboot finished  take a look into /var/log/vzctl.log and /var/log/messages you may do it in two separte ssh sessions with tail -f command.
 
Once reboot finished  take a look into /var/log/vzctl.log and /var/log/messages you may do it in two separte ssh sessions with tail -f command.
  
update vzctl vzquota ploop for better workaround.
+
update vzctl vzquota ploop prlctl for better workaround.
 +
 
 +
**NOTE** If you see messaged like this during module update:
 +
 
 +
  Cleanup    : vzkernel-2.6.32-042stab090.5.x86_64                                                                                                                                                                                          3/4
 +
dkms: removing: sysdig 0.1.84 (2.6.32-042stab090.5) (x86_64)
 +
-------- Uninstall Beginning --------
 +
Module:  sysdig
 +
Version: 0.1.84
 +
Kernel:  2.6.32-042stab090.5 (x86_64)
 +
-------------------------------------
 +
  Status: Before uninstall, this module version was ACTIVE on this kernel.
 +
Removing any linked weak-modules
 +
  sysdig-probe.ko:
 +
  - Uninstallation
 +
  - Deleting from: /lib/modules/2.6.32-042stab090.5/extra/
 +
  - Original module
 +
  - No original module was found for this module on this kernel.
 +
  - Use the dkms install command to reinstall any previous module version.
  
 +
depmod...
 +
DKMS: uninstall completed.
 +
  ------------------------------
 +
Deleting module version: 0.1.84
 +
completely from the DKMS tree.
  
 +
Depending on your system config most likely you need to follow the instructions in the output message and try to install deprecated module manually after reboot.
 +
In general it should not be critical , however it depends on exact module.
 +
In the example we provided you may simply install sysdig in easy way:
 +
1.Install kernel dev with the same version as your vzkernel.
 +
2.Try to install sysdig package that will try to install sysdig-node kernel module.
 +
Otherwise you may find another solutions of module installiation.
  
 
[[Category:Admin area]]
 
[[Category:Admin area]]

Latest revision as of 22:33, 30 April 2020

  • UPD this also may be applied to openvz7 / virtuozzo7
  • ADD consider to complete yum update
 {yeahh the full update} even if you updated only kernel and noted weird errors while starting the CTs or/adn VMs - try to complete full yum update after and restart vz/qemu service even w/o reboot *(as reboot required only for kernel update apply)

Update only vzkernel on Centos. Read till the end before proceed.

Check the bootlist:

cat  /boot/grub/grub.conf

if you using newer grub:

cat /boot/grub2/grub.cfg 

Proceed with update:

yum update vzkernel

Check the bootlis again and note that new kernel appeare to load at the next boot:

cat  /boot/grub/grub.conf

For any case count how much active vps now:

vzlist | wc -l

You may save the output in order to be able to compare it with the same command output after reboot/all vm start

vzlist | wc -l >/home/vzlist.txt
vzlist >>/home/vzlist.txt 

for virtuozzo use:

prlctl list >/home/prlctl.txt

to check/save only openvz7/virtuozzo7 VMs

prlctl list --vmtype vm | wc -l >/home/prlctl_listVMS_wc-l.txt

Reboot:

reboot

Once reboot finished take a look into /var/log/vzctl.log and /var/log/messages you may do it in two separte ssh sessions with tail -f command.

update vzctl vzquota ploop prlctl for better workaround.

    • NOTE** If you see messaged like this during module update:
 Cleanup    : vzkernel-2.6.32-042stab090.5.x86_64                                                                                                                                                                                           3/4 
dkms: removing: sysdig 0.1.84 (2.6.32-042stab090.5) (x86_64)
-------- Uninstall Beginning --------
Module:  sysdig
Version: 0.1.84
Kernel:  2.6.32-042stab090.5 (x86_64)
------------------------------------- 
 Status: Before uninstall, this module version was ACTIVE on this kernel.
Removing any linked weak-modules
 sysdig-probe.ko:
 - Uninstallation
  - Deleting from: /lib/modules/2.6.32-042stab090.5/extra/
  - Original module
  - No original module was found for this module on this kernel.
  - Use the dkms install command to reinstall any previous module version.
depmod...
DKMS: uninstall completed.
 ------------------------------
Deleting module version: 0.1.84
completely from the DKMS tree.

Depending on your system config most likely you need to follow the instructions in the output message and try to install deprecated module manually after reboot. In general it should not be critical , however it depends on exact module. In the example we provided you may simply install sysdig in easy way: 1.Install kernel dev with the same version as your vzkernel. 2.Try to install sysdig package that will try to install sysdig-node kernel module. Otherwise you may find another solutions of module installiation.