Attention is currently required from: iedemam, laforge. neels has posted comments on this change. ( https://gerrit.osmocom.org/c/osmo-bsc/+/27081 )
Change subject: New stats for lchan life duration. ......................................................................
Patch Set 11:
(5 comments)
Patchset:
PS11:
Hi all, […]
for a struct osmo_time_cc *foo, you can use foo->total_sum. It holds the total sum in microseconds (1e-6).
osmo_time_cc doesn't use rbtree, will have to look closer to identify the problem
PS11: (just some comments, haven't fully reviewed yet)
File src/osmo-bsc/bts.c:
https://gerrit.osmocom.org/c/osmo-bsc/+/27081/comment/e851e96c_94ff084e PS8, Line 1514: "Cummulative number of active milliseconds on TCH chans",
Done
you're right, it's not the average indeed
File src/osmo-bsc/lchan_fsm.c:
https://gerrit.osmocom.org/c/osmo-bsc/+/27081/comment/ba625ea8_c9c082b3 PS8, Line 1810:
This is a great bit of information. I do think it is important to capture the error timeouts. […]
another idea would be to have a separate accumulator for the release process (from established_onleave to unused_onenter). it should yield a fairly constant value in normal operation for all lchans. it would be easier to see chan release problems on a separate metric.
https://gerrit.osmocom.org/c/osmo-bsc/+/27081/comment/de81d689_797e7b5c PS8, Line 1811: /* Add active milliseconds to cummulative counts per channel type */
Adopted. […]
ack