mbrestore script doesn't work
Hi,
Just tried to follow the restore steps outlined here (http://techzone.couchbase.com/wiki/display/membase/Membase+Server+versio...) but having tried a few combinations I still can't seem to be able to add the data into a live node..
Basically I'm trying to take backups of one node and then restore them into another node with different IP BUT exactly the same bucket configuration (name, password, size, membase/memcached setting).
Some context:
- I'm using Windows 32-bit Amazon instances
- backup files are generated by running the mbbackup script on one live node (v1.6.5 Membase)
- on the node which I'm trying to restore the data into I did a fresh install of v1.6.5 Membase and configured the buckets to mirror that on the node the backup data is taken from
- bucket I'm trying to restore have just over 2000 items, and the total size of all xx-n.mb files for this bucket is 52MB
- on the new node, ran mbrestore script with -add flag and nothing seems to happen (I waited for at least half an hour and there's next to zero activity on the CPU/Memory as far as I can tell from looking at task manager..)
- tried running mbrestore script without the -add flag, setting the username with the bucket name and setting the password, but none of this seem to make any difference..
Is there anything I'm doing wrong? There is this one entry on the web console's log page:
Control connection to memcached on 'ns_1@xxxxxxxxxxx' disconnected: {{badmatch,
{error,
closed}},
[{mc_client_binary,
stats_recv,
4},
{mc_client_binary,
stats,
4},
{ns_memcached,
ensure_bucket_config,
4},
{ns_memcached,
handle_info,
2},
{gen_server,
handle_msg,
5},
{proc_lib,
init_p_do_apply,
3}]} (repeated 1 times)
which I'm not sure if it's related to my attempts to restore data, is there some moxi log I can look at to figure out what's going on here?
Thanks,
Hi Perry,
We just tried the same exercise again and the restore process worked when we ran the script exactly the same way as before but on a 1.6.5.3 cluster instead. So whatever it was, looks like the new version has fixed it :) !
I wonder though, if it's related to issue MB-3360, although I did use the mbbackup script to generate the files which I then try to restore with.
Anyhow, thanks for getting the new release out!
That error doesn't look particularly related to anything with the restore process.
-Can you verify that the new cluster is able to accept data at all? A simple 'telnet' test should confirm that.
-Can you share the command-line that you're using to run mbrestore?
Perry
Forum support is great for free but sometimes you need a guaranteed response time and dedicated resources for your questions or issues.
Consider purchasing enterprise-level support from Couchbase: http://www.couchbase.com/products-and-services/overview
Call or email "sales -at- couchbase-dot- com" today!