MySQL Forums
Forum List  »  NDB clusters

Re: Impossible to recover from datanode down
Posted by: Jan-willem van Eys
Date: August 30, 2006 04:48AM

Network and disk IO seemed to be marginal.

When I did an all status, I got the response that the starting node was in phase 5.
I tried waiting (overnight) but this didn't resolve it.
I also tried killing the stuck ndbd and restarting it, but that didn't fix anything.

Unfortunately our test servers are in high demand, so I have to request a period for MySQL cluster testing, after which the machines get reallocated to other projects.
This means that I can't just jump in and re-run my tests whenever I feel like it...

I'll request another few days to do a final test, because although I'm getting frutstrated, I can't believe MySQL is as fragile as it currently seems.

What happens if another (the other) node goes during this 40 minutes? How much downtime would it cost to get things back up again, and how much data would be lost?

TIA,
Jan-willem.

Options: ReplyQuote


Subject
Views
Written By
Posted
Re: Impossible to recover from datanode down
1018
August 30, 2006 04:48AM


Sorry, you can't reply to this topic. It has been closed.

Content reproduced on this site is the property of the respective copyright holders. It is not reviewed in advance by Oracle and does not necessarily represent the opinion of Oracle or any other party.