If you are having problems connecting to any services here, please let me know. We are having some peering issues with Comcast causing some issued for Customers accessing us through Comcast. It seems that peering is going through CenturyLink’s network. I’ve generated tickets for this from each end, but the more information we can collect on how widespread this is the better.
If you are having issues, if you can go to our website and under “Web Apps”, pulldown menu, click on Your IP Address and let me know the result. This will help us isolate the failing network node involved. Thank you.
The kernel upgrade last Friday introduced bugs worse than it fixed resulting in some instability. As a result, we’ll be doing another upgrade this Friday to fix the bugs the last bug fixes introduced. Hopefully better results this time, only time will tell.
Reboots are completed. All machines up and verified proper NIS and NFS mounts and all subsystems started successfully. Two machines are still on 5.16.14, those are the UUCP machine and manjaro, and this because I’m having difficulties getting 5.16.17 to build properly on those boxes. I have a bug report opened on bugzilla regarding failed compilation on these machines.
I am going to be rebooting all the machines to upgrade the kernels tonight starting at 11PM. I expect all to be done by 11:30PM. Interruption of individual services should be brief.
Per a customers request, tmux is now installed on all shell servers.
All disks are now good, no more errors, and loads on all machines now are satisfactory, switching to virtio drivers on all the virtual hosts made a HUGE difference. Not hard to get results like below with STATIC HTML content, but find another site that loads PHP based WordPress site in this time frame!
Doing some benchmarking on virtual machines, I discovered on fast drives that virtio drivers are about 10% faster than emulated SATA drives for the SAME hardware.
I had erroneously assumed SATA would be faster because it supported hdparm read-ahead where virtio does not support this parameters, but it turns out that the Linux kernel has demand read-ahead that detects sequential reads and automatically reads ahead and prefills buffers when it detects this. That nullifies this advantage and thus the reduction in interrupts and system calls of virtio provides better performance.
As a consequence, I am going to take the web, mail, and uucp servers down briefly around 11pm tonight to change it’s virtual disks from SATA to virtio.
The brief interruption of web services at 5PM Pacific Time today March 20th was for the purpose of increasing the memory allocated to the web server from 64GB to 80GB to provide a bit more overhead for load spikes.
Ok, running good now that I switched the emulation so going to take down around 11pm’ish to backup as it is. Also going to take Ubuntu down to move to a different physical server. Ubuntu and the web server are the two heaviest loads so doesn’t make good sense to have them on the same physical machine.
Regarding the web server, I discovered the Ethernet device was set to an emulated device, but by switching to virtio, I was able to get the interrupt load down substantially, so going to watch it run for a while and make a decision later with regards to whether or not to move to physical hardware now.