Woke up on Wednesday morning to this…
Not a huge amount of snow, but certainly enough to turn everything white. There was also a fair bit of ice around – saw several cars and a van slide around the road a bit whilst on my way to drop Harry at school.
Month: February 2005
Happy Birthday Lori
Happy Birthday to my wife Lori who, as I write this, is currently on her way back from a talk at the Tate Modern in London and should be arriving back in Nottingham in about 10 minutes.
A new server
I’ve got a new server to replace this one, the primary reason being that with all the sites and services on here there’s less disk space available than would be nice. There’s also the fact that the server is a little slow at times, and is running a really ancient version of RedHat. I’ve tried upgrading it (in place, without a reboot) to Fedora Core 2 and although it works, it’s got a lot of stuff left lying around that shouldn’t be there.
Another disk failure
Well, once again I’ve got a failed disk in my Linux software RAID array. That’s got to be the third one that’s been flagged as failed within the last 6 months – they’re Maxtor 160GB drives. The last one that failed had a comprehensive PowerMAX test run on it and it passed everything. PowerMAX is the DOS utility that you have to run in order to get a special diagnostic code so that you can RMA the drive if it’s still under warranty.
Comment spam again
After having seen at least 12 spam comments added in the last couple of days I finally decided to do something about it. As it’s a fairly widespread problem, there are plenty of pages explaining what to do about it. Some of them are easy to implement but easy for spammers to get around them if they put their minds to it. Others are similarly easy to implement (because someone else has done all the hard work) but would provide a bit more protection.
MySQL Replication problem
Had some problems with MySQL replication yesterday. I setup a new slave to join a bunch of existing ones and found that although ‘LOAD DATA FROM MASTER’ worked fine, once it had got an up-to-date copy of the data that was all it ever had. None of the new transactions from the master were processed on the new slave server.
Solution: server-id on the new slave was set to 1 and conflicted with one of the other servers. Doh!