• Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 515
  • Last Modified:

Unable to access AIX5.2 server due to more virtual memory usage ?

Due to 90% swap space usage We are unable to access the server ? while try access server through telnet or ssh  it is getting exited like below.
$ telnet xxxxxxxxxx
Connected to xxxxxxx.
Escape character is '^]'.
Local flow control off
Connection closed by foreign host.

if I try to access the server through console we are end up with following error
gh space
        Could not fork: Not enough space
                                        Could not fork: Not enough space
                                                                        Could not fork: Not enough space
                                                                                                        Could not fork: Not e

1 Solution
Kerem ERSOYPresidentCommented:
How critical is this server? Can you restart it ? Can you enter in maintenance mode and expand the volume? It seems to me that it is not only a problem with swap and it seems your /sys volume is also topped.
First solution to try:
  • reboot (don't start any applications)
  • increase swap using 'smitty chps' or 'chps -s [# partitions] [paging space name]'
If a normal reboot doesn't bring the system back to a normal state:
  • shutdown
  • boot in maintenance mode
  • increase swap as described above
If swap utilization continues to grow:
  • examine your system for 'runaway' processes:
    • use 'topas, press 'P'. move cursor to 'PAGE SPACE'
  • consider upgrading to 5.3 in order to get rid of possible bugs
KeremE: What is a /sys volume?

Kerem ERSOYPresidentCommented:
Opps sorry :) I meant /usr
Free Tool: Site Down Detector

Helpful to verify reports of your own downtime, or to double check a downed website you are trying to access.

One of a set of tools we are providing to everyone as a way of saying thank you for being a part of the community.

probably is a PS problem but, suppose that the problem is not in PS neither in /usr, but probably in root fs (/) or /var.

If you can, try to login in console, with another user rather than root, and perform a df:
see %Used and from this give an su (not "su -" ).
Change dir to / and remove log file and sure a core file.


Kerem ERSOYPresidentCommented:
I agree with dolomiti. Though it is obvious that it is a filesize problem which is very common with aix.  It might be any critical directory. But if it is preventing the user from loging probably one of /var, / or /usr.

Your only alternative is try to boot in maintenance mode and do a
df -k
Then expand the volueme in quistion. If the problem is with /var you might like to trim down some logs and keep your /var logs under conftrol with built in log removal rules or an external applications like logrotate.

Kerem ERSOYPresidentCommented:
Since this spave issue prevents you from loging maintenance mod will allow you boot since it wont need yo use your /var.

Who's next to throw the dice?
rammaghentharAuthor Commented:
        I am unable to access server. In our setup HMC GUI mode is not working.
How reboot server and go to maintenace mode through HMC command line?
Can I increase paging spce when server is in maintenance  mode ? will smitty chps run in maintenance mode ?
are you even unable to start the GUI from the original HMC terminal? If so, regardless of your actual problem, I'd strongly suggest to fix that issue!
Anyway -
As first steps you should simply reboot the partition and then try to access it using a virtual terminal.
1) reboot - Log in to the HMC as hscroot and issue
chsysstate -r lpar -m managed_system -o shutdown --immed --restart -n lparname
The values for managed-system and lparname are to be supplied by you.
2) open a virtual terminal to the lpar - there are two ways. Either issue at the HMC
mkvterm -m managed_system -n lparname
With vtmenu you will be presented selection panels from which to choose the managed system and the lpar.
Once the terminal has opened, try to log in as root and see what you can achieve.
Ater exiting from the lpar, try to escape the VT session by typing ~~ or .~
Attention - in some cases escaping will not work and your ssh session to the HMC will become unusable. Simply close your session to the HMC and open a new one.
Please post the results of the above. If there was no success, we must go to maintenance mode. Do you have an optical drive which you can attach to the lpar and appropriate AIX installation media? If not, we'll have to use NIM.
Kerem ERSOYPresidentCommented:
Who's next to throw the dice?

I don't know about throwing a dice but I can tell an arrogant person on sight :)
You have to get single session to server - like from console.
Then renice your shell
Then smitty wlm, and cage runaway processes.output of uptime, vmstat 5 5 and oslevel -r would be handy
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

Join & Write a Comment

Featured Post

Keep up with what's happening at Experts Exchange!

Sign up to receive Decoded, a new monthly digest with product updates, feature release info, continuing education opportunities, and more.

Tackle projects and never again get stuck behind a technical roadblock.
Join Now