Attention is currently required from: neels, laforge. iedemam 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:

(2 comments)

Patchset:

PS11: Hi all,

Here is my first stab at converting to use osmo_time_cc. It has only been tested minimally as I wanted some feedback on the approach first.

The previous timestamp+diff method is still used to see per lchan activation ages in the VTY and the new time_cc timekeeper to track and report cumulative milliseconds during channel use. Maybe a new API for osmo_time_cc to pull the elapsed time value could help? Then I would no longer need to use two sources of truth.

There is also a SEGV in here pretty regularly with the following backtrace. Probably a misuse on my part of osmo_time_cc. Just reporting for completeness.

``` Program received signal SIGSEGV, Segmentation fault. rb_insert_color (node=node@entry=0x76bf60, root=0x7ffff7fe2828) at rbtree.c:77 77 rbtree.c: No such file or directory. (gdb) (gdb) bt #0 rb_insert_color (node=node@entry=0x76bf60, root=0x7ffff7fe2828) at rbtree.c:77 #1 0x00007ffff72db95f in __add_timer (timer=0x76bf60) at timer.c:62 #2 osmo_timer_add (timer=timer@entry=0x76bf60) at timer.c:85 #3 0x00007ffff72db9b8 in osmo_timer_schedule (timer=0x76bf60, seconds=0, microseconds=0) at timer.c:107 #4 0x00007ffff72dbcf6 in osmo_timers_update () at timer.c:269 #5 0x00007ffff72dc7b5 in _osmo_select_main (polling=0) at select.c:394 #6 0x00007ffff72dc846 in osmo_select_main_ctx (polling=<optimized out>) at select.c:455 #7 0x000000000040dd55 in ?? () #8 0x00007ffff62cd391 in __libc_start_main () from /lib64/libc.so.6 #9 0x000000000040e7ca in ?? () ```

File src/osmo-bsc/bts.c:

https://gerrit.osmocom.org/c/osmo-bsc/+/27081/comment/0c6222f9_b8425161 PS8, Line 1514: "Cummulative number of active milliseconds on TCH chans",

sounds great

Done