[
https://issues.apache.org/jira/browse/CRAIL-93?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16785720#comment-16785720
]
David Crespi commented on CRAIL-93:
-----------------------------------
Well… it’s most likely I’m still confused 😊
1. “…but all IO requests are done directly from the client to the NVMf
target…” by client, in this case, the namenode correct? All the subsystems
within SPDK can be on the same IP/Port, it’s the Host NQN and Host ID’s that
identify the clients. I guess it could be different if connecting directly to
the remote NVNe devices, but it seems it should work the same way… same options
and all. “…was only possible if those subsystems also had different NQNs.”
All subsystems will have different NQN’s for sure, but each subsystem can have
the same host NQN/host ID assigned. So it sounds like what you are saying is
that Crail is attempting to take the responsibility of creating the subsystems
by creating the subsystem NQN’s?
1. The subsystems have the option of setting the connection by host NQN/
host ID or by not setting that specifically and instead setting the option to
“allow any host”, which opens the subsystem up to discovery by any client, and
any client can connect to it. I’ll have to verify this again, but my belief is
if the subsystem is set this way, a client that is using it host NQN will still
attach to the subsystem… Crail does not. I will verify this behavior today
with a standard client and update the thread.
1. Thanks for the link and suggestions. I’ve watched and reviewed soooo
many of these already… at the end of the day, as you know, we rely on the
documentation, not what someone shows in a presentation, or a talk. The
document within Crail is somewhat lacking, and I’ve also found inconsistencies.
I absolutely know, however, how demanding creating great documentation is, and
that your project is still getting it’s legs, so bare with us with all the
“somewhat” confused questions. I did read about the logging, but I wasn’t sure
it was what we were looking for. I’ll play around with it as well.
1. I assume you meant the “-e”? I’m sorry, I shouldn’t have added the whole
thing… it was just easier to copy and paste it. The -e is for launching a
docker container with that variable and sticking it into the container’s
environment. Nearly all the options that can be applied for setup with Crail,
I pass in to the container environment this way, it makes it so much easier to
alter the behavior. And just to have a complete answer, the variable
“nqn.2017-06.io.crail:cnode”, is your default subsystem NQN that Crail
generates 😊
Regards,
David
> Using Crail with NVMf, the Default NQN also attaches the port number to the
> name.
> ---------------------------------------------------------------------------------
>
> Key: CRAIL-93
> URL: https://issues.apache.org/jira/browse/CRAIL-93
> Project: Apache Crail
> Issue Type: Bug
> Affects Versions: 1.2
> Environment: I'm running crail nodes with docker containers, on a
> Ubuntu 18.04 base.
> Reporter: David Crespi
> Assignee: Jonas Pfefferle
> Priority: Critical
>
> This is the version I'm actually using: v1.1-2-gf0afadc
> I'm set up to use spdk on the backend to crail. When attempting to attach,
> it appears that crail needs two subsystems to achieve the connection. Instead
> of allowing the default name, I have the system variable set: -e
> NVMF_NQN="nqn.2017-06.io.crail:cnode"
> 1) subsystem NQN: nqn.2017-06.io.crail:cnode
> 2) subsystem NQN: nqn.2017-06.io.crail:cnode4420
> 19/03/01 17:46:45 INFO crail: CrailHadoopFileSystem fs initialization done..
> 19/03/01 17:46:45 INFO crail: Connecting to NVMf target at Transport address
> = /192.168.2.104:4420, subsystem NQN = nqn.2017-06.io.crail:cnode4420
> It appears that the initial connect/discovery of the subsystem uses #1, but
> using the
> crail commands (crail fs -mkdir /test) uses #2.
> Both have to have a valid namespace attached as well.
>
> It also appears that when using my own subsystem NQN (NVMF_NQN) name, crail
> wants to generate its own Host NQN. A new one every time. First, how do you
> learn of that
> NQN, and 2nd, it would be great to disable it if spdk has "allow any hosts"
> set. It refuses
> to connect to spdk.
>
>
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
