Thanks. The 2gb memory might be an issue. Lets hope this FS is a little more robust. I've set a cron job to shut down bitcoind each night, and rsync chain state and database dirs. Then restart.
If in the event of a power cut, Im hoping i can just rsync back.
Topic: Power cuts corrupting bitcoin core db (Read 2084 times)
I personally have only used it on BSDs. Here is a write-up and overview on Linux: https://clusterhq.com/blog/state-zfs-on-linux/
A better starting point for installation looks to be: https://github.com/zfsonlinux/pkg-zfs/wiki/Ubuntu-ZFS-mountall-FAQ-and-troubleshooting
A better starting point for installation looks to be: https://github.com/zfsonlinux/pkg-zfs/wiki/Ubuntu-ZFS-mountall-FAQ-and-troubleshooting
Derp, I'm a dummy. I meant ZFS. (if it was ReiserFS I probably wouldn't have mistyperemembered)
http://docs.oracle.com/cd/E19253-01/819-5461/gavvx/index.html
No tip for you!http://docs.oracle.com/cd/E19253-01/819-5461/gavvx/index.html
Trying to follow this article (Ubuntu 14 server), getting the following error. Any ideas? No really hits on google.
http://arstechnica.com/information-technology/2014/02/ars-walkthrough-using-the-zfs-next-gen-filesystem-on-linux/1/
ubuntu@xxx:~$ sudo apt-get install ubuntu-zfs
sudo: unable to resolve host TenThousandAndOne
Reading package lists... Done
Building dependency tree
Reading state information... Done
E: Unable to locate package ubuntu-zfs
Derp, I'm a dummy. I meant ZFS. (if it was ReiserFS I probably wouldn't have mistyperemembered)
http://docs.oracle.com/cd/E19253-01/819-5461/gavvx/index.html
http://docs.oracle.com/cd/E19253-01/819-5461/gavvx/index.html
XFS with checkpoints every hour, then besides having a robust journal that can be replayed when powering up, you can roll back to a previous checkpoint.
Hi Deepceleron,
I've formatted the drive with Blockchain and coinstate on them to xfs. I cant find any documentation on "checkpoints". Can you point me to some or otherwise?
XFS with checkpoints every hour, then besides having a robust journal that can be replayed when powering up, you can roll back to a previous checkpoint.
Thanks for the tip. Looking at xfs now.
XFS with checkpoints every hour, then besides having a robust journal that can be replayed when powering up, you can roll back to a previous checkpoint.
UPS is being looked at, just ads cost and complexity. Looking for software methods first.
I've had a few power cuts on an ubuntu box running bitcoind which requires me to re-index the blockchain. Is it possible to shut down bitcoind each night, backup the chain state and block height, then restart bitcoind? Then if a power failure occurs, I can just restore last good chain state?
Potentially, yes; look into using rsync or a filesystem with checkpoints for your backups. However, depending on the severity and duration of the power cuts, it may be easier to buy a UPS to allow your ubuntu box to keep running through a power failure (or failing that, at least provide it with plenty of time to shut down safely).
I've had a few power cuts on an ubuntu box running bitcoind which requires me to re-index the blockchain. Is it possible to shut down bitcoind each night, backup the chain state and block height, then restart bitcoind? Then if a power failure occurs, I can just restore last good chain state?
Jump to: