The shell server, debian.eskimo.com, will be going down for approximately 45 minutes for maintenance. The virtual machine will be imaged with all the latest updates applied so that if a restore becomes necessary it will be restored to a current condition.
Things will be some what slower than normal today as images made last night are being compressed. This would normally happen over night, but due to the problems encountered with one server last night, the compression is being done during the day today. I’ve niced the jobs down as much as possible to minimize their effect upon other operations.
Everything is back in service.
I am going to have to continue maintenance tomorrow. The shell servers debian, mint, ubuntu, scientific, and scientific7 will be taken out of service for about 20-45 minutes each.
The following servers are up: mail, www, ftp, mail, mx1, shellx, uucp, and centos7.
The following are up but the host is sick: debian, mint, scientific7, scientific, ubuntu, and mx2. These are on the host machine that presently is responding only to ping. The virtual machines are responding but I can’t get into the host, even from the virtuals.
I am going to have to go to the co-location facility and reboot this machine and it’s not going to be graceful. So it would be good not to be doing anything important on this second group of machines until this is completed.
After much gnashing of teeth, I was able to regain access and control of the host that scientific7 sits on remotely so maintenance is proceeding as planned.
The host machine that hosts a number of servers wedged on me during imaging of scientific7. Because of this it is going to be necessary for me to go to the co-location facility. Some of the work scheduled for tonight may have to be completed tomorrow depending upon what all is wrong.
I have a lot of maintenance work to do today and this evening that involves updates, taking systems down for imaging, reboots to make a new libc and openssl active, and reboots of the main file servers to make various updates in them active.
Most of the heavy work that will impact customer access to mail, web, etc, will happen after midnight, first the update of the main file servers and then taking mail, ftp, and the more heavily used shell servers down for imaging. The less used servers that at the time have nobody logged as well as servers that are replicated will be serviced earlier.
It is suggested that you do not plan any uninterruptible activities between midnight and 1AM on 6/6/2015 as this is the time frame the main file servers will be rebooted and during that time all services will be unavailable for a period of approximately 20 minutes.
The shell server, “fedora.eskimo.com“, is back online, now as Fedora 22. There are still some operations in progress like the conversion of the yum database to dnf, the new package handler [Probably stands for D)oes N)ot F)unction].
I really hate it when developers make major changes to what was a good functional system. Anyway it is what it is. Those of you wanting to get a look at Fedora 22, there it is.
The shell server, fedora.eskimo.com, is presently down for an upgrade from Fedora 21 to Fedora 22. If you need a modern RedHat based shell server presently, please use either centos7.eskimo.com or scientific7.eskimo.com.
Also tested for FREAK, not vulnerable to that attack as well.
Both of these vulnerabilities exploit capabilities for browser to fall-back to less secure “export” encryption and are mostly an issue with old browsers.
On this site, we don’t support these flawed encryption standards but you can protect yourself on other sites that do by upgrading your browser to the current version. Those of you stuck with old versions of Explorer because Microsoft didn’t make newer versions available for older operating systems can either obtain a patch from Microsoft or use an alternative browser like Firefox.