And the server went kabloom…
To all of our tens of thousands of loyal readers, I apologize about the outage yesterday.
I’ve promised myself I wouldn’t write about technology, but let’s just say that Solaris 10 x86 UFS filesystem logging doesn’t exactly work as advertised. Combine that with a brownout and a dead battery in the server’s UPS, and you’ve got a day spent running dozens of fsck commands.
Thanks to a bit of creative hacking, coupled with some mighty fine insight into Solaris Zones by The K, we were able to get things all recovered by about 2 AM. Another healthy nod to JeffW, who’s an invaluable resource in all things Solaris.
Apologies if you were roasting a pig and at a loss for what to do during the outage.
Segnala presso:
I too have had problems with Solaris 10 logging. Thanks for sharing, and now I know how to roast a pig, too!
[...] One of the largest problems that I have noticed when doing directory deployments, are either invalid, or non-existent backups of their directory information. They often assure peace of mind when things unexpectedly go wrong (a la unexpected power outages). [...]