clustershell
clustershell copied to clipboard
cea-hpc
Scalable cluster administration Python framework — Manage node sets, node groups and execute commands on cluster nodes in parallel.
Low priority, but possible improvement when there are many group sources (we tend to have more and more group sources, currently 16 on Sherlock, 12 in yaml file + 4...
When parsing a very big nodeset with thousand of nodes like ``` pattern = "nova1000,nova1205,nova1235,nova1001,..." ``` ``` NodeSet(pattern) ``` is much slower (x3-x4) than ``` NodeSet.fromlist(pattern.split(",")) ``` This is even...
Reported during testing of 1.7.82 yaml file: ```yaml 111: foo: 'bar' ``` ``` $ nodeset -f @111:foo nodeset: Unknown group source: "111" ``` ``` $ nodeset -L | grep 111...
In #349 the wildcard support for nodes was added. I've noticed that the constructor of `ParsingEngine` has an optional `node_wildcard_enable` parameter that defaults to `True`. Would you consider a PR...
Multi dimensional nodeset is slower with CS v1.8 compared to CS 1.7.3. Between 25 and 30% slower depending on the operation. ~~~ TESTNAME v1.7.3 v1.8 NodeSet2DContains 0.29597 0.35862 21% NodeSet2DIter...
Related to #260. Would be great to have weighted routes, to either spread the load accordingly, or even able to set up failback routes in topology.conf (eg. a weight of...
I am investigating if we can use "nodeset" instead of our homegrown "nodes" tool (https://www.nsc.liu.se/~kent/nodes/). I lack one feature that we use a lot with nodes --- the possibility to...
A couple of method and classes are deprecated in ClusterShell. We need to add call to `warnings.warn` where needed. The next release will be able to drop this code.
Related to #232. Old signatures (eg. `ev_read(self, worker)`) should be deprecated in 1.x (x > 8) and removed in 2.0.
Utility to avoid gateway reconnections when using clush in scalable mode. This agent is optional. The idea of '''clush-agent''' is that it is started in a login session on the...