Change in osmo-mgw[master]: add DLCX command statistics to osmo-mgw

This is merely a historical archive of years 2008-2021, before the migration to mailman3.

A maintained and still updated list archive can be found at https://lists.osmocom.org/hyperkitty/list/gerrit-log@lists.osmocom.org/.

Neels Hofmeyr gerrit-no-reply at lists.osmocom.org
Thu Nov 8 16:24:41 UTC 2018


Neels Hofmeyr has posted comments on this change. ( https://gerrit.osmocom.org/11521 )

Change subject: add DLCX command statistics to osmo-mgw
......................................................................


Patch Set 4: Code-Review-1

(4 comments)

https://gerrit.osmocom.org/#/c/11521/4/src/libosmo-mgcp/mgcp_protocol.c
File src/libosmo-mgcp/mgcp_protocol.c:

https://gerrit.osmocom.org/#/c/11521/4/src/libosmo-mgcp/mgcp_protocol.c@a1510
PS4, Line 1510: 
rate_ctr_group_alloc() may return NULL also for invalid naming, so I would rather keep these.


https://gerrit.osmocom.org/#/c/11521/4/src/libosmo-mgcp/mgcp_protocol.c@116
PS4, Line 116: 	[MGCP_DLCX_FAIL_NO_RTP] = {"dlcx:no_rtp", "no rtp stream associated with connection."},
I don't think this makes sense from a user perspective ... it's more of an internal error that some part of the endpoint isn't configured properly ... (continuing to comment where it is used)


https://gerrit.osmocom.org/#/c/11521/4/src/libosmo-mgcp/mgcp_protocol.c@1356
PS4, Line 1356: 		     num_conns, ENDPOINT_NUMBER(endp));
this above change is completely orthogonal. I'd argue to keep it in a separate patch, by general principle. Well, ok, just keep it in mind next time... (says the guy with a recurring tradition of dropping humungous code bombs)


https://gerrit.osmocom.org/#/c/11521/4/src/libosmo-mgcp/mgcp_protocol.c@1372
PS4, Line 1372: 		rate_ctr_inc(&rate_ctrs->ctr[MGCP_DLCX_FAIL_NO_RTP]);
IIUC when we don't have this kind of conn, then the endpoint shouldn't exist. MGCP wise: after CRCX, a conn exists right away. After the DLCX on the last conn, the endpoint is gone. So if this happens it's an internal error, more like FAIL_NO_CONN. In practice it will never count up. Right, dexter?



-- 
To view, visit https://gerrit.osmocom.org/11521
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings

Gerrit-Project: osmo-mgw
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: Ie0dde2faf02fd68a69f986973d39b1bea367039b
Gerrit-Change-Number: 11521
Gerrit-PatchSet: 4
Gerrit-Owner: Stefan Sperling <ssperling at sysmocom.de>
Gerrit-Reviewer: Harald Welte <laforge at gnumonks.org>
Gerrit-Reviewer: Jenkins Builder (1000002)
Gerrit-Reviewer: Neels Hofmeyr <nhofmeyr at sysmocom.de>
Gerrit-Reviewer: Pau Espin Pedrol <pespin at sysmocom.de>
Gerrit-Reviewer: Stefan Sperling <ssperling at sysmocom.de>
Gerrit-Comment-Date: Thu, 08 Nov 2018 16:24:41 +0000
Gerrit-HasComments: Yes
Gerrit-HasLabels: Yes
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osmocom.org/pipermail/gerrit-log/attachments/20181108/1b8fe6fd/attachment.htm>


More information about the gerrit-log mailing list