Thank you Mike. Didn’t see that yet.
> Am 16.01.2019 um 16:57 schrieb Michael Robbert :
>
> Andreas,
>
> Look again. I just looked and a commit to the source code was posted to
> the bug yesterday afternoon. It looks like that patch applies to the
> cgroup plugin. It won't show up until the ne
On 17/1/19 1:24 pm, rapier wrote:
Or is there some known and understood way to do something like this that
might be documented somewhere?
What you're describing sounds a lot like Slurm's cloudbursting facility
where nodes can be defined with a name that Slurm knows but when they
boot they in
Yes, well, the trivial cat-skinning method is to use topology.conf to
describe multiple switch topologies confining each architecture to their
meta-fabric. We use GPFS as a parallel filesystem, and all nodes are
connected, but topology.conf keeps jobs on uniform-architecture collectives.
On Thu, J
I recommend putting heterogeneous node types each into their own patition to
keep jobs from spanning multiple node types. You can also set QoS's for
different partitions and make a job in that QoS only able to be scheduled on
nodes=1. You could also accomplished this with a partition config in
We use topology.conf to segregate architectures (Sandy->Skylake), and also
to isolate individual nodes with 1Gb/s Ethernet rather than IB (older GPU
nodes with deprecated IB cards). In the latter case, topology.conf had a
switch entry for each node.
It used to be the case that SLURM was unhappy wi
I don’t actually know the answer to this one, but we have it provisioned to all
nodes.
Note that if you care about node weights (eg. NodeName=whatever001 Weight=2,
etc. in slurm.conf), using the topology function will disable it. I believe I
was promised a warning about that in the future in a
> On Jan 17, 2019, at 4:49 PM, Prentice Bisbal wrote:
>
> From https://slurm.schedmd.com/topology.html:
>
>> Note that compute nodes on switches that lack a common parent switch can be
>> used, but no job will span leaf switches without a common parent (unless the
>> TopologyParam=TopoOptional
And a follow-up question: Does topology.conf need to be on all the
nodes, or just the slurm controller? It's not clear from that web page.
I would assume only the controller needs it.
Prentice
On 1/17/19 4:49 PM, Prentice Bisbal wrote:
From https://slurm.schedmd.com/topology.html:
Note that
From https://slurm.schedmd.com/topology.html:
Note that compute nodes on switches that lack a common parent switch
can be used, but no job will span leaf switches without a common
parent (unless the TopologyParam=TopoOptional option is used). For
example, it is legal to remove the line "Switch
Hi, I'm still in the process of understanding slurm but I was hoping
someone might have an answer for this.
We've a pretty specific use case where we want to have slurm nodes
connected to a slurm controller via VPN. The IPs are assigned to nodes
via OpenVPN.
So is possible to define nodes wi
Nevermind. This was a layer 8 problem. I was editing the wrong
slurm.conf. We recently switched to using RPMs, and I was accidentally
edited the file in the location used before we switched to using RPMs.
It turns out those errors were always there in slurmctld.log, and no one
ever noticed. Now
It appears that 'slurmd -C is not returning the correct information for
some of the systems in my very heterogeneous cluster.
For example, take the node dawson081:
[root@dawson081 ~]# slurmd -C
NodeName=dawson081 slurmd: Considering each NUMA node as a socket
CPUs=32 Boards=1 SocketsPerBoard=4
Our clustername does not have a "-" but the hostname does. Does the slurmdbd
accounting try to create a table name based on the hostname or clustername only?
-SS-
-Original Message-
From: slurm-users [mailto:slurm-users-boun...@lists.schedmd.com] On Behalf Of
Matthew BETTINGER
Sent: Thu
Not Sure if this is related but we ran into an issue configuring accounting
because our clustername had a '-' in the name . This is an illegal character
for table names in mariadb, or used to be.
On 1/17/19, 11:07 AM, "slurm-users on behalf of Sajesh Singh"
wrote:
Trying to setup acco
Trying to setup accounting using the MySQL backend and I am getting errors from
the slurmctld and slurm tools when trying to interact with the accounting
database. Tried starting in debug as well, but could not see anything else that
could point to what could be causing this issue. I have follow
15 matches
Mail list logo