Skip to content

RUNBOOK

BlindBlocArt-Jerry edited this page Sep 12, 2018 · 4 revisions
Before doing and maintenance you should run unregproducer to take your node out of any current schedule
or change key to your backup block producer
#####PLEASE#### run teclos system unregprod {PRODUCER NAME} -p  ####PLEASE###########

!Remember to run regprod after maintenance!

# Maintenance mode on your API/FULL node
This is not as critical, hopefully you have a second node that is load-balanced and just remove the node you need to work on from pool.


# Bring up a new node
After proper installation of current code you can either transfer a backup from you current FULL node and start from there, or use ./start.sh --delete-all-blocks --genesis-json ./genesis.json --hard-replay-blockchain to bring it up to speed.
Remeber to let it get synced up before trying any commands.



# Backups
If you have SSD you can do a snapshot on your FULL node (you should also do this on every node so you have a local copy)
put the following into a file named backup.sh
<
<   #!/bin/bash


<   ./stop.sh
<   cp -r blocks blockstemp
<   cp -r state statetemp
<   ./start.sh
<   rm backup.tar.gz.1
<   mv backup.tar.gz backup.yar.gz.1
<   tar czvf backup.tar.gz blockstemp statetemp
<   rm -rf blockstemp
<   rm -rf statetemp
> (remove all the < & >)
you will have to run chmod +x backup.sh and then you can run it once an hour or so. I run this on backup units so it does not interfer with production, and you can FTP or RSYNC to the production unit if you need to rebuild it (after brining up backup producer). If you had it automated via cron job remmeber to stop that cron job if you have put this backup server as primary.

You can’t perform that action at this time.