Attention is currently required from: jolly, manawyrm, tnt.

laforge has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-e1d/+/34554?usp=email )

Change subject: Automatically reset RIFO on underrun/overflow ......................................................................

Patch Set 3:

(3 comments)

Patchset:

PS3: see detailed comments. what do others think?

also, I think such resync events should be counted in a rate counter.

File src/octoi/e1oip.c:

https://gerrit.osmocom.org/c/osmo-e1d/+/34554/comment/79fdcc40_9034c19f PS3, Line 274: iline_ctr_add(iline, LINE_CTR_E1oIP_E1T_OVERFLOW, 1); I'm a bit worried that we do this after the first packet with out of range frame nr. so a single late or duplicated packet can cause a full reset here. It might be worth considering to wait for several such packets or a certain rate of them?

File src/octoi/octoi_srv_vty.c:

https://gerrit.osmocom.org/c/osmo-e1d/+/34554/comment/d6959207_cc88c233 PS3, Line 419: void set_frames_per_fifo_str(void) why do we have those runtime generated strings? can't we generate them at compile time? We're passing a constant/define only