Feature #1820


don't open E1/T1 timeslots not used by the BSC

Added by laforge over 7 years ago. Updated over 7 years ago.

Start date:
Due date:
% Done:


Spec Reference:


As far as I know, libosmo-abis / openbsc currently open all timeslots of a E1/T1 interface, even those not actually needed/used.

We should make sure that the code really only opens those timeslots (sockets in mISDN, devices in DAHDI) that the BSC needs.

This is a pre-condition for a BSC-located PCU, as that PCU will want to open some of those timeslots for PCU TRAU frames.

Actions #1

Updated by laforge over 7 years ago

  • Target version set to RBS2000 with BSC-located PCU
Actions #2

Updated by laforge over 7 years ago

at least for DAHDI it seems this is already the case, so this issue can actually be closed, I presume:

root@sysmobsc-e1:~# cat /proc/dahdi/1
Span 1: TE4/0/1 "T4XXP (PCI) Card 0 Span 1" ESF/B8ZS 

           1 TE4/0/1/1 Clear (In use) 
           2 TE4/0/1/2 Clear (In use) 
           3 TE4/0/1/3 Clear (In use) 
           4 TE4/0/1/4 Clear 
           5 TE4/0/1/5 Clear 
           6 TE4/0/1/6 Clear 
           7 TE4/0/1/7 Clear 
           8 TE4/0/1/8 Clear 
           9 TE4/0/1/9 Clear 
          10 TE4/0/1/10 Clear 
          11 TE4/0/1/11 Clear 
          12 TE4/0/1/12 Clear 
          13 TE4/0/1/13 Clear 
          14 TE4/0/1/14 Clear 
          15 TE4/0/1/15 Clear 
          16 TE4/0/1/16 Clear 
          17 TE4/0/1/17 Clear 
          18 TE4/0/1/18 Clear 
          19 TE4/0/1/19 Clear 
          20 TE4/0/1/20 Clear 
          21 TE4/0/1/21 Clear 
          22 TE4/0/1/22 Clear 
          23 TE4/0/1/23 Clear 
          24 TE4/0/1/24 Clear 
root@sysmobsc-e1:~# lsof | grep -i dahdi
osmo-nitb  649     root    5u      CHR              249,1      0t0       8588 /dev/dahdi/chan/001/001
osmo-nitb  649     root    6u      CHR              249,2      0t0       8589 /dev/dahdi/chan/001/002
osmo-nitb  649     root    7u      CHR              249,3      0t0       8590 /dev/dahdi/chan/001/003

When activating a Um timeslot for PDCH, we could instruct the ICP to switch it to a completely different E1/T1 timeslot.

Actions #3

Updated by laforge over 7 years ago

  • Status changed from New to Resolved
  • Assignee changed from dexter to laforge
  • % Done changed from 0 to 100
Actions #4

Updated by dexter over 7 years ago

Also checked it - looks good:

root@sysmocom:~# lsof | grep "/dev/dah" 
osmo-nitb 19904           root    4u      CHR              249,1      0t0       9353 /dev/dahdi/chan/001/001
osmo-nitb 19904           root    5u      CHR              249,2      0t0       9363 /dev/dahdi/chan/001/002
osmo-nitb 19904           root    6u      CHR              249,3      0t0       9365 /dev/dahdi/chan/001/003
Actions #5

Updated by laforge over 7 years ago

  • Status changed from Resolved to Closed

Also available in: Atom PDF

Add picture from clipboard (Maximum size: 48.8 MB)