Nobody has used MxLinux this month at all. They are preparing a new release but it’s based upon an antique 4.19 kernel and an antique 1.20 version of Mate, and consequently I’ve decided rather than upgrading the machine I am just going to retire it and load some other version of Linux on it. If anyone WANTS to use MxLinux say something now before it goes away. If anyone has a favorite they’d like to recommend this is a good time to make your suggestions to support@eskimo.com.
We had a brief outage around 7:35PM caused by one of our physical host servers spontaneously rebooting. I do not know what caused the reboot as there was nothing logged of the event.
We had a brief problem with webmail which resulted from my failure to add our own IP space to the ignoreip line of the fail2ban configuration in the new mail server. This has been fixed.
Because an upgrade exploded and left Ubuntu in a bad state, I need to restore it from backup and start over. It will be down for about an hour.
I’m going to attempt an upgrade of the web server to Ubuntu 19.10 tonight. This will include an upgrade to PHP 7.3, however, this is likely to break a few things initially. If web mail is not working late tonight, be patient. Typically PHP makes syntactical changes that requires the upgrading of some applications but it is impossible to know which in advance.
So what I am going to do is take it down for about half an hour this evening to make a current backup in case I can not get all the applications to work with 7.3 and we need to back out. Like most PHP upgrades, newer versions are faster and therefore preferable if we can get all the applications to work.
Machine is again running with full RAID.
The failed drive has been replaced. The RAID array is copying to the new drive, it will take about ten hours to complete.
I will be taking most services down around midnight for approximately two hours to replace a failed hard drive on the server that provides NFS service for everyone’s home directories.
Maintenance completed. Mail is available.
I am taking the client mail server down for about 1/2 hour to image it after getting smartlists working again.