Suse /Boot Partition nealry full

SUSE Linux Enterprise Server SLES Configure-Administer
1

The /boot partition on SLES 11.1 server is 92% full. I created a rather small one based on the recommendation of Yast and some white papers I happened to read. This is whats currently in my boot partition.

[HR][/HR]
-rw------- 1 root root 512 2012-06-27 15:12 backup_mbr
lrwxrwxrwx 1 root root 1 2012-07-09 12:27 boot → .
-rw-r–r-- 1 root root 1236 2012-10-15 16:11 boot.readme
-rw-r–r-- 1 root root 108117 2012-04-27 19:31 config-2.6.32.59-0.3-default
-rw-r–r-- 1 root root 108998 2012-04-27 18:34 config-2.6.32.59-0.3-trace
drwxr-xr-x 2 root root 1024 2012-07-09 12:33 grub
lrwxrwxrwx 1 root root 26 2013-08-06 22:52 initrd → initrd-2.6.32.59-0.3-trace
-rw------- 1 root root 9590589 2012-07-03 14:58 initrd-2.6.32.12-0.7-trace-kdump
-rw------- 1 root root 9722142 2012-06-27 15:16 initrd-2.6.32.36-0.5-default-kdump
-rw-r–r-- 1 root root 7114153 2013-08-06 22:52 initrd-2.6.32.59-0.3-default
-rw------- 1 root root 10187620 2013-08-06 22:52 initrd-2.6.32.59-0.3-default-kdump
-rw-r–r-- 1 root root 7114063 2012-07-09 17:23 initrd-2.6.32.59-0.3-default.old
-rw-r–r-- 1 root root 6986380 2013-08-06 22:52 initrd-2.6.32.59-0.3-trace
-rw------- 1 root root 9699337 2013-08-06 22:53 initrd-2.6.32.59-0.3-trace-kdump
drwx------ 2 root root 12288 2012-06-27 15:01 lost+found
-rw------- 1 root root 435712 2012-06-27 15:12 message
-rw-r–r-- 1 root root 200557 2012-04-27 19:34 symsets-2.6.32.59-0.3-default.tar.gz
-rw-r–r-- 1 root root 206105 2012-04-27 19:58 symsets-2.6.32.59-0.3-trace.tar.gz
-rw-r–r-- 1 root root 203720 2012-04-27 20:44 symsets-2.6.32.59-0.3-xen.tar.gz
-rw-r–r-- 1 root root 497257 2012-04-27 19:33 symtypes-2.6.32.59-0.3-default.gz
-rw-r–r-- 1 root root 502751 2012-04-27 19:57 symtypes-2.6.32.59-0.3-trace.gz
-rw-r–r-- 1 root root 487604 2012-04-27 20:43 symtypes-2.6.32.59-0.3-xen.gz
-rw-r–r-- 1 root root 179424 2012-04-27 19:33 symvers-2.6.32.59-0.3-default.gz
-rw-r–r-- 1 root root 180774 2012-04-27 19:52 symvers-2.6.32.59-0.3-trace.gz
-rw-r–r-- 1 root root 1664799 2012-04-27 19:26 System.map-2.6.32.59-0.3-default
-rw-r–r-- 1 root root 1971633 2012-04-27 19:37 System.map-2.6.32.59-0.3-trace
-rw-r–r-- 1 root root 3804083 2012-04-27 19:31 vmlinux-2.6.32.59-0.3-default.gz
-rw-r–r-- 1 root root 4235179 2012-04-27 19:50 vmlinux-2.6.32.59-0.3-trace.gz
lrwxrwxrwx 1 root root 27 2012-07-09 12:33 vmlinuz → vmlinuz-2.6.32.59-0.3-trace
-rw-r–r-- 1 root root 3256832 2012-04-27 19:26 vmlinuz-2.6.32.59-0.3-default
-rw-r–r-- 1 root root 3612288 2012-04-27 19:37 vmlinuz-2.6.32.59-0.3-trace
[HR][/HR]

I’m running Kernel 2.6.32.59-0.3 I believe so some of these files can go. I’m also wondering if I can remove the “kdump” files as well. I can’t patch until I have more room. On the next servers I build I’m making /boot much bigger diskspace is fairly cheap but time is not.

2

Hi sirguy,

I’m running Kernel 2.6.32.59-0.3 I believe so some of these files can go.

yes, you can manually delete the files referencing the older kernels. Sometimes it happens when a newer kernel is installed, sometimes it doesn’t :frowning:

Regards,
Jens

3

Thanks Jens. Any idea about those “kdump” files?

4

Hi sirguy,

I assume you mean i.e. “initrd-2.6.32.12-0.7-trace-kdump”: Those are (as all the other initrd files) the contents of the “initial RAM disk”, which are loaded automatically when the according kernel is loaded. The “-kdump” versions are created since you have the “kdump crash kernels” configured - a special version of the kernel that is loaded if your regular kernel detects a “kernel crash” and is used to create i.e. analysis data. This is probably the result of installing the “trace” kernel or configuring kdump support, i.e. via “yast kdump”.

In short: Only delete the kdump initrd files for such kernels that aren’t available in /boot. If you want to get rid of them completely, remove kdump support from your configuration, else those initrds will be automatically recreated with every “mkinitrd” invocation (i.e. kernel update).

Regards,
Jens