Abstract
When I tried to apply bruco to recent locked data, I found bruco server couldn't access data.A problem seemed to be happen after the Kashiwa planned power outage.
So I fixed a connection problem between the data server and the bruco server.
They are now online.
Details (for server maintainer)
Bruco server accesses to data via NFS@k1nds2 not via NDS@k1nds2 from the view point of a speed of data access. The entity of the data on k1nds2 is the full data of Kashiwa. This implementation is for reducing an access load to Kamioka DAQ disks.After the planned power outage at Kashiwa, connection from k1nds2 to Kashiwa (see also k1nds2:/etc/fstab) was automatically recovered. So the access to the past data via NDS@k1nds2 had been available on various tools such as ndscope, diaggui etc. without any recovery works. But nfs-kernel-server service couldn't still found the data after recovering the connection between k1nds2 and Kashiwa. On the other hand, NFS client on bruco server could access the NFS@k1nds2. So bruco process stopped during the data access and it never came back.
This issue could be fixed by restarting nfs-kernel-server service on k1nds2 and re-mounting NFS region on the bruco server. I'm not sure how to recover it in automatically. For now, these recovering procedures must be done in manually after a recovery from the disconnection between Kamioka and Kashiwa.