Manjaro will be down for a while.  It is hurt to the point where a fresh install is easier than straightening out dependency loops.

     Depending upon weather and other factors, I am planning to upgrade the kernel 6.0.11, however this depends upon the roads being clear of snow.  And then after the kernel upgrade, I am planning to go to the co-location facility and will be making some adjustments to the server that is serving the majority of virtual private servers.  It is currently slightly unstable so I am going to adjust the clock down from 4.3 Ghz to 4.2 Ghz to provide improved stability.  This will probably happen around 1AM.

     The kernel upgrades will affect all services, and should be completed by around 11:30PM.  The server clock adjustment will affect only virtual private servers and should be relatively brief at around 1AM.  This will just be changing a multiplier from 43 to 42, so just need to take it down, adjust, and reboot.

     The services affected by the kernel upgrade will include all of Eskimo’s paid services including shell accounts, e-mail, web hosting, etc, and free federated services including https://friendica.eskimo.com/, https://hubzilla.eskiimo.com/, and https://nextcloud.eskimo.com/ and https://yacy.eskimo.com/.

     I need a referral for someone who repairs torch down roofs.  I’ve got a branch that came down and pierced my roof and having difficulty finding a contractor that does torch down in residential and isn’t totally swamped already because of the storm in the Seattle area, I am in Shoreline.

     I’d rather give insurance company money to someone in the community than not.  If you do this kind of work or know someone who does, please contact me at nanook@eskimo.com or 206-812-0051.  Thank you.

     Ok, I found that I had X11forwarding set to No in ssh_config.  I actually did not but something changed it, probably a Ubuntu upgrade.  Setting that to Yes made it work again.

     But now here is the deeper mystery, there is something in my .home directory also affecting this, because even when set to No, if I started with a clean directory it worked, if I ssh -X to another machine it worked, or if I ssh -X to another account on eskimo.com, it worked.

     So it would seem something in my home directory overrides the ssh_config settings and made it work and that got blitzed.  But I don’t know what that is.  So I am functional again but really would like to have a deeper understanding.

     I’m looking for some community wisdom here, particular in light of being totally overwhelmed between roof issues, four days of power outages, water bed failures, dryer failure, heater failure, I’ve obviously done something to piss-off the creator.

     Anyway, I use the old eskimo.com system still as the NIS master and also do some of the system accounting using some x-apps.  To do this I use X redirection via ssh by connecting ssh -X eskimo.com.

     The home directory is NOT out of space or quota, yet I get an error that says xauth can not create the .Xauthority file.  But if I ssh to another account on that machine it works fine so it has to be something in my home directory environment that is messed, yet, ssh -X to any other machine works so it also has to be something specific to this machine.  I am wondering if I am running into some file system limit of SunOS but of course it is not giving any errors that would help.

     Any ideas would be much appreciated.

     Owing to recent events and the fact that another snow storm is forecast for Friday evening and 6.0.10 seems to have introduced a file descriptor leak, I am putting off kernel upgrades until December 9th starting at 11pm by which time hopefully 6.0.11 will be out and addressed this newly introduced issue.

     I apologize that I have been unresponsive for the last three days.  I have been without electricity or an Internet connection and now have a tree limb sticking through my roof and ceiling.

     We just got power back after a three day outage, one of the very last of City Light’s out of service customers to be restored.

     So I am working through the backlog and general maintenance.  Please have patience, I will get to your messages and needs as soon as possible.

     Last night when we had a server lock up owing to a kernel bug, there was an unforeseen consequence discovered today.

     It appears the mail server was bound to this server as the NIS server it was using for authentication.  This caused authentication to fail during this interval.  If you attempted to access the mail server too many times during this period, your IP address may be locked out now.

     If you are presently unable to access the mail server, in order to unblock you, I need to know your device IP address for the device or devices that can not access mail.

     To find this out, FROM the affected device go to our website https://www.eskimo.com/, hover over “Web Apps“, and in the pull-down menu that results, select “Your IP address” from the bottom.  Let me know what that IP address is then I can investigate and unblock you if you are blocked.

     Since the file server that crashed contained both home directories, a lot of the web applications, and MariaDB database, it’s sudden refusal to continue screwed up a lot of things which I’ve spent the entire evening fixing, so those of you who made payments today will see your receipts sometime tomorrow.  I apologize for the delay but I’m just too burned out to continue tonight.

     If there is a silver lining it is that that server and all the virtual machines on it now have the new kernel installed so won’t need to do those on Friday.  Also other machines that I needed to reboot to resolve NFS issues, etc, are also updated, so Friday’s upgrades should go faster than normal and have a relatively low chance of failure since failures almost always involve this particular machine both because it has more cores than the other and thus more parallel processes with potential to trip race conditions.