Change in osmo-ttcn3-hacks[master]: f_channel_needs_modify(): ask actual TCH lchan mode via VTY
gerrit-no-reply at lists.osmocom.org
Tue Apr 20 14:40:26 UTC 2021
neels has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-ttcn3-hacks/+/23807 )
Change subject: f_channel_needs_modify(): ask actual TCH lchan mode via VTY
Patch Set 1:
> Patch Set 1:
> I mean, we shouldn't rely on what osmo-bsc tells us about it, since we are actually testing it. We should know from the test what to expect without asking. Perhaps passing g_pars.expect_channel_mode_modify there or similar would help?
I actually disagree the way f_establish_fully() currently does all sorts of magic decisions like this.
I would much more appreciate a test that expects what should happen 1:1 and not assume/infer important parts of the test in the way f_establish_fully() does.
But I am not going to completely refactor f_establish_fully(), we're using that everywhere and it is likely going to cause quite some effort.
I'm also not going to duplicate and modify that code, because I'd have to also duplicate a lot of the channel activation testing code and, as ugly to me as it is, it is apparently working quite well for the last years.
The quickest way to achieving a working Mode Modify test is asking the BSC for the current channel mode.
At first I thought, indeed, we could also use that mad f_rslem_get_last_act() to recall the last channel activation message and read the mode from that. But in fact we would also have to check whether the last activation was ACKed or NACKed, whether another Channel Mode Modify happened in the meantime and changed the mode, and whether the channel was released or not. So, directly asking the BSC for the current actual mode is the most accurate and least error prone way to keep using f_establish_fully().
To view, visit https://gerrit.osmocom.org/c/osmo-ttcn3-hacks/+/23807
To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings
Gerrit-Owner: neels <nhofmeyr at sysmocom.de>
Gerrit-Reviewer: Jenkins Builder
Gerrit-CC: pespin <pespin at sysmocom.de>
Gerrit-Comment-Date: Tue, 20 Apr 2021 14:40:26 +0000
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the gerrit-log