07-02-2003
Please post your OS and version, any error messages from /var/adm/messages (or if you have it configured other than default - from the file you are sending error messages to), the output from df -k, version and type of any volume manager you are using.
The following probably won't help right off since you state you don't have any free space.
From Sunsolve:
Quote:
Problem: receiving error message:
WARNING: /tmp: File system full, swap space limit exceeded
/tmp becomes full when the following quasi equation looks like this for your
system:
((ram + swap - processes) * .9) - files_in_/tmp = 0
| | | | |
| ---------- | /tmp (avail)
| | |
| virtual memory |
| |
-------------------------------
|
/tmp (total)
Resolution Top
There are many things that can be done to remedy the problem, among them;
1. add more ram
2. add more swap
3. kill processes
4. delete files in /tmp
A fast and simple solution to add more swap is to create a swap file
and add it to the existing swap:
#mkfile <megs>m <path_to_filesystem_with_plenty_of_space>/swap
#swap -a <path_to_filesystem_with_plenty_of_space>/swap
Example:
#mkfile 20m /newdir/swapfile
#swap -a /newdir/swapfile
To make this swap space available (add the swap file to virtual memory) each
time the system is rebooted, add a line in the /etc/vfstab.
#device device mount FS fsck mount mount
#to mount to fsck point type pass at boot options
...
/swapfile - - swap - no -
This is inefficient though. Filesystems waste space for administrative reasons
(about 10%). They are slower than accessing raw disk too.
It is better to use a raw disk parition as a swap file.
There may be plenty of virtual memory. The problem could be that some processes
are simply using too much swap and need to be killed/restarted. Some badly
written/configured programs will take up all available memory no matter how
much of it is available.
The see which processes are using the most memory run the following command:
# /usr/bin/ps -el | sort -rn -k 10
8 S 52475 2594 345 0 40 20 60b1adc0 17325 6098872e ? 57:24 Xsun
8 S 52475 8111 8092 0 51 20 60d6efa0 3882 61915dfe pts/4 1:06 sotool
8 S 52475 9054 19313 0 51 20 60d07658 1568 pts/6 0:26 dtmail
8 S 0 3211 1 1 43 20 609f2038 1207 60989a16 ? 66:41 esd
8 S 0 3213 1 0 40 20 6095c6e0 1146 60989a66 ? 27:01 esd
8 S 52475 2739 2728 0 51 20 60cea208 1128 60b2a0c6 ? 0:01 nametool
^
The above ps output is sorted by process size. |
The biggest offenders are at the top. |
The 10th column is the process size in pages. _______|
To see how big a page is in kbytes, run the pagesize command:
# pagesize
8192
10 More Discussions You Might Find Interesting
1. HP-UX
Hi All,
one of the mount point in Hp ux server has reached 95%
its a data base file and can not be deleted.
so i want to know how to increase the size of mount point
i am new to unix ,please help me (1 Reply)
Discussion started by: jyoti
1 Replies
2. Solaris
Hi Guys
I need to increase the size of my /tmp swap file. What is the easiest way to do this.
Thanks
Carson (2 Replies)
Discussion started by: cmackin
2 Replies
3. UNIX for Dummies Questions & Answers
Dear all,
I wanna to extend my /tmp size. I deleted the all files of /tmp but it still shows there's no more space on it.
The OS is Red Hat Enterprise Linux AS release 4 (Nahant Update 5).
# df /tmp/ -h
Filesystem Size Used Avail Use% Mounted on
/dev/mapper/VolGroup00-LogVol00... (3 Replies)
Discussion started by: mr_bold
3 Replies
4. UNIX for Dummies Questions & Answers
Hi All,
Could anybody tell me how to increase/decrease a process size in UNIX HP machine ? Whether the process size limitation will cause core dump ?
Thanks,
Rohit.. (1 Reply)
Discussion started by: ronix007
1 Replies
5. Solaris
Hi all
system Solaris 10 10/09 s10x_u8wos_08a X86
ufs file system
I would like to install SunStudio.
After the gui comes up it shows that i need to add swap space of at least 900m.
the command swap-s shows 880m free.
My question is can you temporarily increase swap by 1 or 2... (3 Replies)
Discussion started by: kc2dws
3 Replies
6. UNIX for Dummies Questions & Answers
hi guys
I am working on my vmware workstation.
I have a /dev/sdb which is 5GB. I am using LVM.
Now I increase /dev/sdb 2 more GB.
fdisk -l shows 7 GB but pvscan still shows 5GB.
how do I make my system recognize the new 7GB added and be able to add those to my physical volumen and... (1 Reply)
Discussion started by: kopper
1 Replies
7. Solaris
Dear All,
How to increase the swap size when physicall memory reaches 60 %. OR it can be only done after the physicall memory is full.
Rgds
Rj (8 Replies)
Discussion started by: jegaraman
8 Replies
8. Solaris
Hi,
the /tmp size is less whereas the size allocated to swap is quite big. how to increase the size of /tmp -
#: swap -l
swapfile dev swaplo blocks free
/dev/md/dsk/d20 85,20 8 273096 273096
#: swap -s
total: 46875128k bytes allocated + 2347188k reserved =... (2 Replies)
Discussion started by: psb74
2 Replies
9. Solaris
Hello All,
I have solaris server running,
uname -a
SunOS host 5.9 Generic_112233-12 sun4u sparc SUNW,Sun-Fire-280R
Filesystem Size Used Available Capacity Mounted on
/dev/md/dsk/d0 9.8G 8.7G 1.0G 90% /
/dev/dsk/c1t1d0s3 4.3G 7.7M 4.2G ... (17 Replies)
Discussion started by: gull05
17 Replies
10. Solaris
Hi
Please let me know how to increase the size of /tmp file system from 512m to 1024m dynamically without reboot in solaris zone
# df -h
Filesystem size used avail capacity Mounted on
zones/zone1 11G 1.0G 10.0G 10% /
/dev 11G 1.0G ... (6 Replies)
Discussion started by: sb200
6 Replies
LEARN ABOUT DEBIAN
xfs_estimate
xfs_estimate(8) System Manager's Manual xfs_estimate(8)
NAME
xfs_estimate - estimate the space that an XFS filesystem will take
SYNOPSIS
xfs_estimate [ -h? ] [ -b blocksize ] [ -i logsize ]
[ -e logsize ] [ -v ] directory ...
DESCRIPTION
For each directory argument, xfs_estimate estimates the space that directory would take if it were copied to an XFS filesystem. xfs_esti-
mate does not cross mount points. The following definitions are used:
KB = *1024
MB = *1024*1024
GB = *1024*1024*1024
The xfs_estimate options are:
-b blocksize
Use blocksize instead of the default blocksize of 4096 bytes. The modifier k can be used after the number to indicate multiplica-
tion by 1024. For example,
xfs_estimate -b 64k /
requests an estimate of the space required by the directory / on an XFS filesystem using a blocksize of 64K (65536) bytes.
-v Display more information, formatted.
-h Display usage message.
-? Display usage message.
-i, -e logsize
Use logsize instead of the default log size of 1000 blocks. -i refers to an internal log, while -e refers to an external log. The
modifiers k or m can be used after the number to indicate multiplication by 1024 or 1048576, respectively.
For example,
xfs_estimate -i 1m /
requests an estimate of the space required by the directory / on an XFS filesystem using an internal log of 1 megabyte.
EXAMPLES
% xfs_estimate -e 10m /var/tmp
/var/tmp will take about 4.2 megabytes
with the external log using 2560 blocks or about 10.0 megabytes
% xfs_estimate -v -e 10m /var/tmp
directory bsize blocks megabytes logsize
/var/tmp 4096 792 4.0MB 10485760
% xfs_estimate -v /var/tmp
directory bsize blocks megabytes logsize
/var/tmp 4096 3352 14.0MB 10485760
% xfs_estimate /var/tmp
/var/tmp will take about 14.0 megabytes
xfs_estimate(8)