I broke the graphical interface on centos7.eskimo.com by installing some 3rd party software that was incompatible. It is fixed now.
The old eskimo.com shell server is revived. I’m pretty much out of spare parts now. If anyone has a working Sun SS-10 they want to get rid of, or good parts for one, please let me know.
What appears to have happened is that three of four heat sink retainers broke and the heat sink separated from the CPU and it cooked.
If you feel like exploring another new server while eskimo.com is down, ‘mint.eskimo.com’ is now online running the Mint 17 Linux distribution.
Please note that the only graphical interface presently supported for this machine is x2go. If you come in with x2go please use the ‘Mate’ desktop. The x2go platform does not support desktops which require accelerated graphics or compositing desktops.
The old Sparc based SS-10 SunOS 4.1.4 shell server, ‘Eskimo.Com’ is down. Eskimo.Com is still the NIS master and thus important. I will be out of the office for probably a couple of hours to work on it. I hope to be back by 3pm or so.
Those of you who have been contemplating Centos 7, we’ve just turned up a new shell server with Centos 7 and it is available for your use although there aren’t a lot of applications installed yet.
Presently the only way to access it graphically is with x2go, and then you must use either Mate or KDE desktop. The Gnome is Gnome 3 and presently doesn’t work without accelerated graphics which remote desktop applications can not provide. However, Mate is a continuation of Gnome 2 and works much as Gnome 2 worked under Centos 6 as used on shellx.
The server name is centos7.eskimo.com. Your same login, password, files, etc, will be present.
Debian.eskimo.com is finished. All servers are restored to service and no additional service affecting outages will be required this weekend.
All machines are returned to service except debian.eskimo.com and it should be back in about fifteen minutes. All customer facing and non-replicated servers maintenance will be completed this morning so there will not be additional outages Sunday morning.
Maintenance is still in progress.
Shellx took much longer than normal because I lost connectivity with the server near the end of the imaging process and had to start over. It will be back up momentarily.
Scientific is also down being imaged. Debian.eskimo.com is still up and available as is the old ‘eskimo.com’ sever.
Shellx is backup now and is finished for the evening.
The mail server will be next and then ftp / www. After scientific completes Debian will be next.
I rebooted the web server at 11:15AM because PHP applications were responding slowly and I was unable to find any heavy load to justify the cause. Since the machine needed to be rebooted to load a new kernel anyway I booted it. After the boot it responded normally. Downtime was approximately two minutes.
I have to apply kernel updates, take the machines down, and image them, and reboot all of the servers. I will be doing this early Saturday morning and if I am unable to complete all of them in time I will extend this to the same period Sunday morning.