Need a VM for development

[johnbent]

The release VM (OVA) that we have currently is awesome for quick deployment and testing.

However we also need a release VM (OVA) that is setup for development. Such a VM should come bundled with all the build dependencies.

Anything else @andriytk ?

[shailesh-vaidya]

Hi @johnbent - Can we use docker image created as part of #89 for this. Developer can setup VM on any platform and follow steps mentioned at #89 (comment) to generate rpm's/artifacts.

Also Developer can use artifacts generated from above steps to perform CORTX deployment on local VM . Provisioner accepts target repo url as file:///var/artifacts/0/ .

[andriytk]

@shailesh-vaidya whatever. The user just have to take OVA, check out the code and be able to build and run the code right away.

Makes sense?

[johnbent]

Hey @andriytk that's what we're working towards. Thanks for the succinct definition of done. @justinzw will be testing the instructions @shailesh-vaidya posted above and if it works (especially if it works just as your succinctly describe), @justinzw will be working to get this documented with @VenkyOS .

@VenkyOS , when we get this documented, here's how I want the links to work:
https://github.com/Seagate/cortx/blob/main/CONTRIBUTING.md#prerequisites -> new_file_in_parent_doc_dir

[VenkyOS]

@johnbent - Noted.

[shailesh-vaidya]

Hello @VenkyOS , @justinzw - Please let me know if you face any issues in testing build instructions located at #89 (comment) to generate rpm's/artifacts.

[shailesh-vaidya]

@VenkyOS - Hope other than timing there no issue. Can you please document steps . It does takes around 45mins to build full stack in Jenkins environment also.

[VenkyOS]

@shailesh-vaidya - Yes. Timing is an issue. I followed your instructions in the document. Shall I document and format the same in RST format? If yes, which repository should it be placed.

[shailesh-vaidya]

@justinzw Could you please validate steps once.

@VenkyOS - I feel we can put this document in https://github.com/Seagate/cortx repo. We are using GitHub Docker registry to store docker images. Hence we can make docker image public without making https://github.com/Seagate/cortx-re public.

Refer - https://github.com/orgs/Seagate/packages/container/package/cortx-build

[shailesh-vaidya]

@justinzw , @VenkyOS Please use make clean build -i for now. This is required due to known issue in cortx-manager build. Its tracked in EOS-16486 Exclude return code check for csm_setup and csm_test.

[VenkyOS]

@shailesh-vaidya - Sure. But, is it fine to showcase the Known Issue.

[shailesh-vaidya]

@VenkyOS Yes. You can showcase it as known issue. CSM team is working on fixing it.

[VenkyOS]

@shailesh-vaidya - Done.

[shailesh-vaidya]

@andriytk @johnbent We have published docker image to build entire CORTX at https://github.com/orgs/Seagate/packages/container/package/cortx-build

Instructions to build CORTX packages using this Docker image is documented at - https://github.com/Seagate/cortx/blob/main/doc/Release_Build_Creation.rst

[andriytk]

Does it allow build and run Motr on the VM?

cc @hessio

[shailesh-vaidya]

@andriytk This docker image supports only build Motr and the other components. Running Motr through it is not possible as of now.

[andriytk]

Then this is not what's needed.

[johnbent]

I asked @hessio to look at this also. @shailesh-vaidya, after building it, how do we run it? That, as @andriytk points out, is clearly the next step and I think @hessio got stuck there as well.

[andriytk]

How to start a singlenode cluster on the VM (Motr+Hare) is described at https://github.com/Seagate/cortx-hare.

[shailesh-vaidya]

Adding details from email thread here

@hessio and me tested single node cluster on the VM (Motr+Hare) is described at https://github.com/Seagate/cortx-hare/blob/main/README.md .

I was able to follow document within Seagate Network . However just want to note few points,

1. lctl list_nids was giving below error. I observed that lnet module was not loaded at https://github.com/Seagate/cortx-hare/blob/main/README.md#prepare-the-cdf stage. I need to load it using modprobe lnet command.

[root@ssc-vm-1613 ~]# lctl list_nids
opening /dev/lnet failed: No such file or directory
hint: the kernel modules may not be loaded
IOC_LIBCFS_GET_NI error 2: No such file or directory

2. make deinstall was failing with error as below. Need to install lbzip2 package using yum install lbzip2 -y to fix this.

tar (child): lbzip2: Cannot exec: No such file or directory
tar (child): Error is not recoverable: exiting now
tar: Child returned status 2
tar: Error is not recoverable: exiting now
make: *** [vendor/dhall-bin/1.26.1/bin/dhall] Error 2

3. At Run I/O test. steps command gives following error. However test worked.

[root@ssc-vm-1613 ~]# /opt/seagate/cortx/hare/libexec/m0crate-io-conf >/tmp/m0crate-io.yaml
Usage: grep [OPTION]... PATTERN [FILE]...
Try 'grep --help' for more information.
Usage: grep [OPTION]... PATTERN [FILE]...
Try 'grep --help' for more information.

4. hctl status showing two unknown services.

[root@ssc-vm-1613 ~]# hctl status
Data pool:
    # fid name
    0x6f00000000000001:0x2f 'the pool'
Profile:
    # fid name: pool(s)
    0x7000000000000001:0x4f 'default': 'the pool' None None
Services:
    localhost  (RC)
    [started]  hax        0x7200000000000001:0x6   192.168.15.233@tcp:12345:1:1
    [started]  confd      0x7200000000000001:0x9   192.168.15.233@tcp:12345:2:1
    [started]  ioservice  0x7200000000000001:0xc   192.168.15.233@tcp:12345:2:2
    _[unknown]  m0_client  0x7200000000000001:0x29  192.168.15.233@tcp:12345:4:1
    [unknown]  m0_client  0x7200000000000001:0x2c  192.168.15.233@tcp:12345:4:2_
Devices:
    localhost
    [M0_NC_UNKNOWN]  /dev/null
    [M0_NC_UNKNOWN]  /dev/loop1
    [M0_NC_UNKNOWN]  /dev/loop2
    [M0_NC_UNKNOWN]  /dev/loop3
    [M0_NC_UNKNOWN]  /dev/loop4
    [M0_NC_UNKNOWN]  /dev/loop5
    [M0_NC_UNKNOWN]  /dev/loop6
    [M0_NC_UNKNOWN]  /dev/loop7
    [M0_NC_UNKNOWN]  /dev/loop8
    [M0_NC_UNKNOWN]  /dev/loop9
    [M0_NC_UNKNOWN]  /dev/loop0

[hessio]

Can we close this issue based on this devvm we have available in  the motr repo https://github.com/Seagate/cortx-motr/releases/tag/ova-centos79

[andriytk]

See also Seagate/cortx-motr#1355.

Here is also CentOS 8 based OVA - https://github.com/andriytk/cortx-motr/releases/tag/centos83ova.

[hessio]

Ok so it is ok to close this issue as we have this requirement?

Or should we wait until Seagate/cortx-motr#1355 is merged?

[cortx-admin]

For the convenience of the Seagate development team, this issue has been mirrored in a private Seagate Jira Server: https://jts.seagate.com/browse/EOS-28332. Note that community members will not be able to access that Jira server but that is not a problem since all activity in that Jira mirror will be copied into this GitHub issue.

[hessio]

Closing as as we have CentOS 8 based OVA - https://github.com/andriytk/cortx-motr/releases/tag/centos83ova
 

and - Seagate/cortx-motr#1355

[cortx-admin]

Patrick Hession commented in Jira Server:

Issue closed

[cortx-admin]

Patrick Hession commented in Jira Server:

Issue closed

[cortx-admin]

Patrick Hession commented in Jira Server:

Issue closed