RE: Just saw - Call for new committers

2021-03-24 Thread Marc
> > I found rexray to be a bit buggy with releasing rbd's on time, making
> it difficult for tasks to move to different hosts. But I tested this
> maybe more than a year ago.
> 
> It's still buggy in that point. But in our case, it does not happen
> often.
> 
> > You can give it a try. Be carefull using the original csi driver, it
> overwrites your /etc/ceph/ceph.conf
> > https://github.com/f1-outsourcing/csiceph/tree/master
> 
> I will try in the next days and give you feedback.
> 
> Can you show me the "mesos csi plugin json" u use for ceph?

[@~]# cat /etc/mesos-csi/csi-ceph.json
{
  "type": "rbd.csi.ceph.io",
  "endpoints": [
{
  "csi_service": "NODE_SERVICE",
  "endpoint": "unix:///tmp/mesos-csi-uuh6Ra/endpoint.sock"
}
  ],
  "target_path_root": "/mnt/mesos-unman-plugin"
}


RE: Just saw - Call for new committers

2021-03-24 Thread Marc
> > Any interest in using csi with ceph for stateful tasks?
> 
> I would! We already use ceph in mesos but via rexray. I was thinking to
> take a deeper look into to csi driver therefore I can remove rexray from
> our stack.
> 
> 

I found rexray to be a bit buggy with releasing rbd's on time, making it 
difficult for tasks to move to different hosts. But I tested this maybe more 
than a year ago.

Because I want to limit linux capabilities on tasks I have decided to configure 
the external/unmanaged csidriver.


[@c04 mesos-csi]# tail /etc/rc.local

# mesos csi nfs plugin
/usr/libexec/csi/csinfs-start.sh
# mesos csi ceph plugin
/usr/libexec/csi/csiceph-start.sh


[@c04 mesos-csi]# cat /usr/libexec/csi/csiceph-start.sh
#!/bin/bash
#

umask 0077

export CSI_DEFUSERID="app.mesos.test"
export CSI_DEFUSERKEY="KE=="

mkdir /tmp/mesos-csi-uuh6Ra 2>/dev/null
cd /tmp/mesos-csi-uuh6Ra
/usr/libexec/csi/csiceph -endpoint 'unix:///tmp/mesos-csi-uuh6Ra/endpoint.sock' 
-nodeid $HOSTNAME -type rbd -drivername ceph-csi-rbd -v 10 
-log_file=./csiceph-plugin.log -logtostderr=false >stdout 2>stderr &

You can give it a try. Be carefull using the original csi driver, it overwrites 
your /etc/ceph/ceph.conf
https://github.com/f1-outsourcing/csiceph/tree/master




RE: Just saw - Call for new committers

2021-03-24 Thread Marc


> I saw that Qian Zhang was calling about new commiters. Well, I already
> ask one month ago on slack if I can help the Mesos project. :-) Benjamin
> told me I should subscribe these Mailinglist, but what should I say, I
> missed the confirmation mail and was wondering why the mailinglist is so
> silent. :-)

Yes it is. Great place to send yourself messages and meditate! :)

> To me; Mesos is a important part of my own company. I drive it for all
> of our own services but I also help customers to install and run it in
> there environment. 

Any interest in using csi with ceph for stateful tasks? It is really nice to 
see that these tasks are able to take their storage to whatever hosts their 
deployed on.
I was asking the developers of the cephcsi a while ago to change their code so 
it actually works as a csi driver and not as a kubernetes driver. And they are 
willing to do some work to create the required functionality for the slrp. 
I currently test a bit with preprovisioned rbd images in tasks via mvp. (I had 
to change code of csiceph driver for that to get it to work)

>I also developed a mesos kafka framework and I'm the
> developer of the "Mesos Executor" (and soon the Mesos Operator) for
> Apache Airflow 2.x. Yeah I know, thats just small things. But I'm
> willing to learn and grow.

You know anything of the Marathon? They implemented csi secrets driver via some 
proprietary secrets solution, not using the default mesos secrets. (ao. that is 
why I had to change the cephcsi driver)

> So... Here I'm and it would be my pleasure to help. :-)
> 
You have my appreciation!