“dmg system stop” should not stop daos_server processes only child daos_io_server processes/ranks. “dmg system stop” on master will issue a SIGTERM to daos_io_server and unfortunately currently handli

commit d7ce4b783ffdc87b33f0c1606276983632a72392 Author: Tom Nabarro <tom.nabarro@...> Date: Sat Apr 4 19:22:25 2020 +0100 DAOS-4164 build: upgrade spdk to v20.01 (#1902) Upgrade spdk, remove spd

The superblock stores allocated ranks: “ [tanabarr@wolf-71 daos_m]$ cat /mnt/daos0/superblock version: 0 uuid: 8359d5d6-1d8f-430c-90db-205fc07113bf system: daos_server rank: 2 validrank: true ms: fals

Thanks for the positive feedback, much appreciated. We are looking to add information such as swim activity and other metrics to the dmg system query capability. Regards, Tom Nabarro – DCG/ESAD M: +44

“dmg -l <access_point_host> system query --verbose” will give you details of ranks that have previously joined the system including current state. This is an area of active development. https://daos-s

Did you do a “git submodule update” to pull in the scons_local changes? Maybe I missed that out of the instructions, there was a change to the configure script check on the spdk libs. Regards, Tom Nab

Non-root supported for SCM only currently. Apologies for any inconsistency with that communication. Why don’t you try building with the PR I suggested, that should get landed before too long. Regards,

Apologies that this has taken some of your time, up until now the only real supported option for running DAOS with NVMe is to run as root (many months ago you could get away with running UIO+SPDK as n

Hello Kevan, sincere apologies for not being able to reply sooner, I haven't been well and Mike has been on leave. SPDK through VFIO (when running as non-root) broke with the version of SPDK (19.04) w

Could you please try with number of hugepages set to 8192 in server config. We are also in the process of updating to SPDK v20.01.1 for DAOS 1.0-1.2 (TBD). Regards, Tom Nabarro – DCG/ESAD M: +44 (0)77

Hello Anton, "daos_server storage prepare -n" should be run prior to "daos_server storage scan" to make the devices visible to DAOS (“daos_server storage prepare -n --reset” will make them visible in

Hello Patrick I'm looking into this now

There seems to be multiple messages about not being able to meet the required number of hugepages requested. I was not involved in the work but there have been changes in how huge page allocation is p

Hello Kevan, Found some time to revisit this, would be good if you had a moment to tryout this PR and comment if it’s resolved the issues you were seeing (should hopefully fix both server and client s

Again thanks for the detailed response, I have pushed https://github.com/daos-stack/daos/pull/1637 and updated the ticket it relates to https://jira.hpdd.intel.com/browse/DAOS-3924 in line with what t

Thanks for the detail, The access points refer to remote hosts which might not be listening on the same port as specified in the host local config file. E.g. hostA has config file which specifies Port

After looking this again, I actually think this is expected. “access_points” in the server config file is indicating to starting daos_server instances where the MS access points are in the system (to

If you want to try a PR that would be great, otherwise I will try to get to it soon. Seems like there is some inconsistency in the client side config arisen from different people trying to do differen

I will look into this, that looks very undesirable. https://jira.hpdd.intel.com/browse/DAOS-3924 Regards, Tom Nabarro – DCG/ESAD M: +44 (0)7786 260986 Skype: tom.nabarro

https://github.com/daos-stack/daos/blob/master/utils/config/daos.yml it’s the client configuration file and if you supply the –debug option dmg will tell you where it’s reading it in from. [tanabarr@w