Paul Millar

Hi Chris, That's certainly interesting work, trying to get btrfs working. My impression is that, a number of years ago, btrfs looked very promising, possibly eclipsing zfs; however, the momentum...

> ZFS always had (and likely always will have) the license issue Yes, I quite agree. Although I'm also surprised how little that seems to matter. AFAIK, it's relatively easy...

Capturing the NFS traffic between the client and server (using a tool like tcpdump) is a good bet. This might (or might not) provide some additional insight into what's going...

OK, quick summary: the client seems to be scanning through the directory response and requesting file attributes (GETATTR) for the contents of the directory. AFAIK, this is normal behaviour for...

Note that doors will push known checksums into the namespace. Therefore this problem is more involved that simply fixing pools. To create a clear separation between expected and actual checksums,...

Could you verify that each of your ZK servers has the same information about the core nodes? To do this, you can use a command like: ``` bin/zkCli.sh -server NODE:2181...

You should find a `.zookeeper` file on the SRM node. The file should have a name like `/var/log/dcache/srm_lcg-lrz-srm.zookeeper`. Could you somehow attach an extract from that file? The extract should...

Thanks for sharing. It doesn't look like there was any problems, at least at the curator level.

Thanks for the info so far. I have another question. All domains should establish a TCP connection to all core domains. This uses port 11111 by default. If a domain...

OK, good. So, the most likely explanation is that the problematic domain (the pool domain, in the latest example) believes there are only two domains. It simply doesn't connect to...