[9fans] Fwd: lib9p: Add clunk callback to Srv struct

[9fans] Fwd: lib9p: Add clunk callback to Srv struct

[Tiago Natel]

[-- Attachment #1: Type: text/plain, Size: 967 bytes --]

Someone here can help me?

---------- Forwarded message ----------
From: Tiago Natel <tiago.natel@neoway.com.br>
Date: 2016-02-01 19:17 GMT-02:00
Subject: lib9p: Add clunk callback to Srv struct
To: 9front@9front.org


Hello folks,

Is there a reason why lib9p doesn't have a clunk function pointer in Srv
struct?

I have a file server project using Srv and I want to know when no one
client have a specific file opened.

One way that I was able to get this working was forking 9front and adding a
clunk callback to Srv structure. See the commit below:

https://bitbucket.org/tiago4orion/plan9front/commits/5e1141f0a4aa98310cb0e2382c5c78c60fe73b4f

My project usage of the clunk routine is here:
https://bitbucket.org/tiago4orion/dchan/src/60dc3e45eb28c8a8289c177680120ef7f44e0925/fs.c?fileviewer=file-view-default#fs.c-680

This makes sense? Or is there better ways to achieve this?
And if that makes sense, it can go upstream?

Thanks.

[-- Attachment #2: Type: text/html, Size: 1698 bytes --]

Re: [9fans] Fwd: lib9p: Add clunk callback to Srv struct

[cinap_lenrek]

> Is there a reason why lib9p doesn't have a clunk function pointer in Srv struct?

what about Srv.destroyfid()?

          Destroyfid
               When a Fid's reference count drops to zero (i.e., it
               has been clunked and there are no outstanding requests
               referring to it), destroyfid is called to allow the
               program to dispose of the fid->aux pointer.

--
cinap

Re: [9fans] Fwd: lib9p: Add clunk callback to Srv struct

[Skip Tavakkolian]

> I have a file server project using Srv and I want to know when no one
> client have a specific file opened.

i believe since you're using alloctree and passing in the destroy function, you don't need it.

Re: [9fans] Fwd: lib9p: Add clunk callback to Srv struct

[Tiago Natel]

[-- Attachment #1: Type: text/plain, Size: 1665 bytes --]

2016-02-01 20:03 GMT-02:00 <cinap_lenrek@felloff.net>:

> > Is there a reason why lib9p doesn't have a clunk function pointer in Srv
> struct?
>
> what about Srv.destroyfid()?
>
>           Destroyfid
>                When a Fid's reference count drops to zero (i.e., it
>                has been clunked and there are no outstanding requests
>                referring to it), destroyfid is called to allow the
>                program to dispose of the fid->aux pointer.
>
>
Thanks for your help! I'd tried using destroyfid to achieve what I need but
failed. I tried today again implement with destroyfid but realized that it
will not fully support what I need.

I'm using a file server for exchange data between 9P clients. When a new
file is created, I create a plan9 channel and two threads (one for handle
reads and other for writes), a write(2) to the file is translated into a
sendp and a read(2) is translated into a recvp on the channel. The channel
could be buffered or not, and then I want to maintain data allocated (aux
related data) anyway, because the file server is a queueing system when
channel have a buffer bigger than zero.

Apart from that, I want to know how many clients have each file opened to
update my stats file. It's possible in any way without a clunk callback?

I'm trying to replace a rabbitmq server with this system, but I have a
requirement for some way of monitoring of queues size, performance of
channels, number of clients connected to each channel (file on dchan), etc,
I need this kind of information for make a comparison with the current
queue system...

Thanks!

--
> cinap
>
>

[-- Attachment #2: Type: text/html, Size: 2311 bytes --]

Re: [9fans] Fwd: lib9p: Add clunk callback to Srv struct

[Skip Tavakkolian]

[-- Attachment #1: Type: text/plain, Size: 1859 bytes --]

i'd recommend studying 9pserve.c in plan9port.

On Tue, Feb 2, 2016 at 10:17 AM Tiago Natel <tiago.natel@neoway.com.br>
wrote:

> 2016-02-01 20:03 GMT-02:00 <cinap_lenrek@felloff.net>:
>
>> > Is there a reason why lib9p doesn't have a clunk function pointer in
>> Srv struct?
>>
>> what about Srv.destroyfid()?
>>
>>           Destroyfid
>>                When a Fid's reference count drops to zero (i.e., it
>>                has been clunked and there are no outstanding requests
>>                referring to it), destroyfid is called to allow the
>>                program to dispose of the fid->aux pointer.
>>
>>
> Thanks for your help! I'd tried using destroyfid to achieve what I need
> but failed. I tried today again implement with destroyfid but realized that
> it will not fully support what I need.
>
> I'm using a file server for exchange data between 9P clients. When a new
> file is created, I create a plan9 channel and two threads (one for handle
> reads and other for writes), a write(2) to the file is translated into a
> sendp and a read(2) is translated into a recvp on the channel. The channel
> could be buffered or not, and then I want to maintain data allocated (aux
> related data) anyway, because the file server is a queueing system when
> channel have a buffer bigger than zero.
>
> Apart from that, I want to know how many clients have each file opened to
> update my stats file. It's possible in any way without a clunk callback?
>
> I'm trying to replace a rabbitmq server with this system, but I have a
> requirement for some way of monitoring of queues size, performance of
> channels, number of clients connected to each channel (file on dchan), etc,
> I need this kind of information for make a comparison with the current
> queue system...
>
> Thanks!
>
> --
>> cinap
>>
>>

[-- Attachment #2: Type: text/html, Size: 2833 bytes --]

Re: [9fans] Fwd: lib9p: Add clunk callback to Srv struct

[cinap_lenrek]

> I'm using a file server for exchange data between 9P clients. When a new
> file is created, I create a plan9 channel and two threads (one for handle
> reads and other for writes), a write(2) to the file is translated into a
> sendp and a read(2) is translated into a recvp on the channel. The channel
> could be buffered or not, and then I want to maintain data allocated (aux
> related data) anyway, because the file server is a queueing system when
> channel have a buffer bigger than zero.

It is hard to say without seeing the code, but this construction sounds wrong
as recvp() in Srv.read would block the 9p read loop causing you to
not process any other 9p requests when one client is blocked in a read.
You also want to handle flushes otherwise you cannot interrupt/cancel
the blocked read. You can handle this by chaining the Req's that you
cannot satisfy immidiately in a linked list and respond to them from some
other proc or thread once you have data the client could read.

What do you mean by client? You could have multiple Srv's with each
client having its own network connection to it. Or it could mean multiple
attaches (mounts) and a single Srv. Or you it could mean you mean *someone*
reading the file and the state about a "Client" is in the Fid. Anyway,
it sounds like the problem you'r having with destroyfid() is that it doenst
give you access to information whoever is the client of the file closed no?

--
cinap

Re: [9fans] Fwd: lib9p: Add clunk callback to Srv struct

[Tiago Natel]

[-- Attachment #1: Type: text/plain, Size: 3080 bytes --]

2016-02-02 23:25 GMT-02:00 <cinap_lenrek@felloff.net>:
>
>
> It is hard to say without seeing the code, but this construction sounds
> wrong
> as recvp() in Srv.read would block the 9p read loop causing you to
> not process any other 9p requests when one client is blocked in a read.
> You also want to handle flushes otherwise you cannot interrupt/cancel
> the blocked read. You can handle this by chaining the Req's that you
> cannot satisfy immidiately in a linked list and respond to them from some
> other proc or thread once you have data the client could read.
>

 Source code: hg clone https://bitbucket.org/tiago4orion/dchan

Well, I'm doing a lot of tests with simultaneous access and everything
'appears' to works as expected.
I'm using threadlistensrv/threadpostmountsrv to handle each new 9P
connection in a separate thread.

When a new client creates a file, the file server creates a Channel and two
additional threads (one for block the reads, and one for block the writes)
that will enqueue (reqqueuepush) the related requests. See here:

Create a new file (or dchan channel): fs.c:497,507

When someone attempts to write to this file: fs.c:/^fswrite
The request will be enqueued in the f->aux->wq (Reqqueue*) executing the
function syncwrite on the writer thread of this file.

The fs.c:/^syncwrite will send the r->ifcall.data in the channel associated
with the file (f->aux->chan) and it will block the writer thread (of this
specific file) until someother 9P client attempts to read the file. The
read is similar, but happens in the syncread function.

That way I'm able to create multiple clients:

plan9:
cpu% mount -c /srv/dchan /n/dchan

Linux:
# mount -t 9p -o port=6666 ip.of.dchan /n/dchan

# cd /n/dchan
# touch pipeline
The command above will create two associated with this specific file.

# cat pipeline
The command above will block in the recvp call in the reader thread of this
file, until someone writes something in it.

cpu% cd /n/dchan
cpu% ls
ctl stats pipeline
cpu% echo AAAAAAAAAAAAA >> pipeline

The command above will issue a sendp in the same channel on the writer
thread. The linux box will receive (recvp) the data and block in the next
read(2).

That's the workflow.

>
> What do you mean by client?


By client I mean a 9P connection: mount -c /srv/dchan /n/dchan


> You could have multiple Srv's with each
> client having its own network connection to it.


I'm using thread*srv function to achieve this. It's correct, isn't?


> Or it could mean multiple
> attaches (mounts) and a single Srv. Or you it could mean you mean *someone*
> reading the file and the state about a "Client" is in the Fid. Anyway,
> it sounds like the problem you'r having with destroyfid() is that it doenst
> give you access to information whoever is the client of the file closed no?
>

 Yes, I don't know if I understodd what you mean, but yeah, I need the
information when someone closed the file (stopped reading the Channel).

Thanks

>
> --
> cinap
>
>

[-- Attachment #2: Type: text/html, Size: 4798 bytes --]