[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: consecutive locks on the same session



John Ng (jng) wrote:

If we decide it is an error, should the session terminate? Or continue and just reply with the error?



We already decided a long time ago it is an error. The section cited below explains what to do if a lock is in use.

Why would we want to terminate the session?
That's a bit severe I think.  It's just an <rpc> failing.

However, in reading prot-06 again, the text has a bug.
It's hard to spot because of the hyphenated word at
the end of the line (ID-nit!)

-----------------------------------

section 7.5, heading 'Negative Response',  para 2,  sentence 1:

OLD:
If the lock is already held, the <error-tag> will be in-use and ...


NEW:

 If the lock is already held, the <error-tag> will be 'lock-denied' and ...

----------------------------------

Note that 'lock-denied' error code exists because
the 'session-id' of the lock holder needs to be returned.
The errors 'in-use' and 'resource-denied' are intended
for generic cases or internal constructs not exposed to users.

--jng




Andy


-----Original Message-----
From: owner-netconf@ops.ietf.org [mailto:owner-netconf@ops.ietf.org] On Behalf Of Steven Berl (sberl)
Sent: Wednesday, June 01, 2005 12:00 PM
To: James Balestriere (jbalestr); netconf@ops.ietf.org
Subject: RE: consecutive locks on the same session


Didn't see any other replies to this, so I will try.

I don't think the current draft specifies this behavior. I propose that an attempt to lock an already locked configuration should generate an error just as if the lock were held by another session. The error message contains the session-id of the session holding the lock. The manager software can compare this session-id with its current session-id (which it received in the hello) and know that the lock is held by itself.

Other opinions?

-steve



-----Original Message-----
From: owner-netconf@ops.ietf.org
[mailto:owner-netconf@ops.ietf.org] On Behalf Of James Balestriere (jbalestr)
Sent: Wednesday, May 18, 2005 3:32 PM
To: netconf@ops.ietf.org
Subject: consecutive locks on the same session



if a session does a lock and it gets the lock we send ok.
if it does a lock again whilst it still has the lock, does


it get an

error or ok ?

I am suspecting we send ok but it is not very clear from the spec.

James.




--
to unsubscribe send a message to


netconf-request@ops.ietf.org with the

word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/netconf/>



--
to unsubscribe send a message to netconf-request@ops.ietf.org with the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/netconf/>





-- to unsubscribe send a message to netconf-request@ops.ietf.org with the word 'unsubscribe' in a single line as the message text body. archive: <http://ops.ietf.org/lists/netconf/>






--
to unsubscribe send a message to netconf-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/netconf/>