I attended my 2nd Spectrum Scale User group meeting, it was amazing to see how much feedback IBM had taken onboard from 2015 and has directly influenced the product. It was also very useful to see how other users are solving their problems using GPFS. All the slides from the conference are online, the rest of this post picks out some specific areas I found particularly interesting and how they could apply to the University of Bristol....
The following post contains my notes from the GPFS User Group meeting in York. Keynote Doris Conti, Director, Spectrum Scale (GPFS) and HPC SW Product Development Doris’ Keynote started the day with a theme that continued throughout the day: IBM are encouraging their users to get in touch with developers to help steer the direction of GPFS. They are also looking for customers to join Beta programs for various components of GPFS....
GPFS 3.3 Introduced License designations, for both client and server nodes. So after upgrading a cluster from GPFS 3.2, you are required to designate licenses with the mmchlicnse command. I recently upgraded a GPFS cluster from 3.2 to 3.5 which contained 6 servers and 393 clients. Unfortuantly mmlslicense does not agree with me and has determined it requires 396 server licenses and 7 client licenses. 1 2 3 4 5 6 7 Summary information --------------------- Number of nodes defined in the cluster: 403 Number of nodes with server license designation: 0 Number of nodes with client license designation: 0 Number of nodes still requiring server license designation: 396 Number of nodes still requiring client license designation: 7 Even using the mmchnode --client did not demote the client....
I experienced the following error on starting up a GPFS node in a cluster: unexpected token conflict in recovery: majType 1 minType 7 tokType Inode key F657ED5089ABDE89:00000000000FC191:0000000000000000 node 1 mode xw flags 0x0 seqNum 1141739 The resulting in the node in question asserting and other nodes remote asserting, making the cluster unstable. IBM states this is related to the token memory (TM) management which can lead to unexpected result when being exhausted....