Tag error: <txp:posted /> ->  : Non-static method timezone::is_dst() should not be called statically  on line 1265
Callandor: Unix minutae
Jan 24, 12:44 PM
Category  

Oh, the problems I run into at work. I do find solving most of them fun though.

Our Dell Poweredge 2650s are starting to die. I had one lose three of the five disks in its RAID 5 array last week. That means the data is unrecoverable, folks. Its firmware was years out of date; it was not until we put the disks in another 2650 that we were told the disks had failed.

Lantronixes are a wonderful thing. Remote secure out-of-band management? Remote console management when a machine’s network/SSH is broken? Yes please. I just wish Sun (and Cisco) didn’t necessitate rolled cat5 cables to hook up serial management ports to them. That’s serial-over-cat5, not straight-through or crossover pinouts. All of the HP and Linux machines I’ve tried need the DB25-to-RJ45 adapter or the DB9-to-RJ45 (of which I have plenty) and a regular ethernet cable, but most of the Sun machines (the important ones, of course) need a rolled cable or adapter, of which I have one. Applies to: Sunfire V445, Sunfire V210, and probably more.

SUSE 10.0’s yast2 installation module is seriously broke. As in, it segfaults when I try to update it, as I had to do a lot of updating to SP1 in the past month (SP0 is no longer supported). Thus leading to RPM hell! “All” you have to do is download the right RPMs for the installation module, all its dependencies, then apply them, and then you can resume updating the system:

sudo rpm -Uvh yast2-installation-2.13.192-0.4.noarch.rpm yast2-packager-2.13.175-0.5.ppc.rpm yast2-pkg-bindings-2.13.117-0.13.ppc.rpm yast2-2.13.104-0.7.ppc.rpm yast2-country-2.13.52-0.9.ppc.rpm yast2-slp-2.13.4-0.17.ppc.rpm yast2-core-2.13.39-0.3.ppc.rpm yast2-inetd-2.13.8-0.6.noarch.rpm libzypp-2.15.10-0.4.ppc.rpm yast2-ncurses-2.13.66-0.12.ppc.rpm yast2-perl-bindings-2.13.11-0.17.ppc.rpm libzypp-zmd-backend-7.1.1.0_0.8-0.16.ppc.rpm

Simple, no?

More whining about error messages: Solaris 10, NFS mounts, “Not owner”? Edit /etc/default/nfs and change NFS_CLIENT_VERSMAX to 3. AIX, NFS mounts, “Permission denied”? Check your domain in resolv.conf (YMMV, my maps rely on the search domain). RedHat, ypserv, ‘Can’t send “Clear” req to local ypserv’? Add an ipv4 localhost line to /etc/hosts.

Last thing you’d ever think to check until now: Got network problems? Make sure your MAC address isn’t all 0s! Some switches don’t like that, that does not belong. I triple-checked all the settings on my (Solaris 8) system, confimed that it was negotiating 100mbps with the switch, but could not get any traffic in or out! Until I changed the card’s MAC address, which took far too long to find. In the process I also learned that Intel Pro/1000 Desktop NICs are not compatible with Solaris on SPARC.. so much for my spare NICs being useful.

Comment

Commenting is closed for this article.