Admin sys

Stuff that happens to sysadmins

Buy one 1U server from $supplier, specifically ask for a pair of rails, learn that « Of course it comes with rails! ». Two weeks later. Buy a 42U rack, and eight 1U servers, all of these from the same $supplier, and at the same time. Receive your rack and 8...

Continuer à lire...

Geek pr0n again

Take one 3U server from SuperMicro, fitted with an Adaptec 31605 SATA RAID controller and 16 one-terabyte disks. Create a RAID 0+1 array for data storage, with 14 of these 16 disks. You get: #available storage [root@sam119 ~]# df -h /data Filesystem Size Used Avail Use% Mounted on /dev/sdb1...

Continuer à lire...

RPM Hell

Actually, this is not RPM Hell but rather crappy packaging hell: [root@sam70 ~]# rpmbuild –rebuild xen-3.1.0-10.fc8.src.rpm Installing xen-3.1.0-10.fc8.src.rpm error: Failed build dependencies: transfig is needed by xen-3.1.0-10.x86_64 libidn-devel is needed by xen-3.1.0-10.x86_64 texi2html is needed by xen-3.1.0-10.x86_64 SDL-devel is needed by xen-3.1.0-10.x86_64 curl-devel is needed by xen-3.1.0-10.x86_64 libX11-devel is...

Continuer à lire...

Stuff that happens to cluster sysadmins

Suddenly getting 60% packet loss between the LAN and the cluster, just because kernel: ip_conntrack: table full, dropping packet. Fix the problem by raising the limit from 65536 to 2097152 slots. If this limit is reached, that’ll eat 700MB of memory, ouch.

Continuer à lire...