OPERATIONAL DEFECT DATABASE
...
...
Under heavy polling of BRIDGE-MIB you might experience MTS buffer leak in recv_q of "SNMP Response SAP" with subsequent crash of SNMP process. Indication of the issue: =================== n5k# sh sys int mts buf sum node sapno recv_q pers_q npers_q log_q sup 12693 4 0 0 0 sup 180 0 1 0 0 sup 284 0 3 0 0 sup 27 146686 0 1 0 <=== !!! 2018 Apr 30 04:00:38.327 n5k %KERN-2-SYSTEM_MSG: [6927862.165180] mts_do_msg_input(): Send Failure due to NO SPACE - node=4, sap=27, uuid=26, pid=23939, sap_opt = 0x1, hdr_opt = 0x0, rq=209585(5 3653760), lq=0(0), pq=0(0), nq=0(0), sq=0(0), fast: rq=0, lq=0, pq=0, nq=0, sq=0, maxl=4096: - kernel 2018 Apr 30 04:00:38.431 n5k %KERN-2-SYSTEM_MSG: [6927862.165184] - kernel 2018 Apr 30 04:00:38.531 n5k %KERN-2-SYSTEM_MSG: [6927862.165188] mts_print_longest_queue_state: opcode counts for first and last 50 messages in recv_q of sap 27: - kernel 2018 Apr 30 04:00:38.575 n5k %KERN-2-SYSTEM_MSG: [6927862.165195] mts_print_msg_opcode_in_queue: opcode 7679 - 100 messages - kernel 2018 Apr 30 04:01:08.629 n5k %KERN-2-SYSTEM_MSG: [6927891.810303] mts_is_q_space_available_haslock_old():2214: regular+fast mesg total = 209564, soft limit = 1024 - kernel 2018 Apr 30 04:01:08.000 n5k %KERN-2-SYSTEM_MSG: [6927891.810310] mts_is_q_space_available_haslock_old(): NO SPACE - node=4, sap=27, uuid=26, pid=23939, sap_opt = 0x1, hdr_opt = 0x0, rq=209564(5 3648384), lq=0(0), pq=0(0), nq=0(0), sq=0(0), fast: rq=0, lq=0, pq=0, nq=0, sq=0, maxl=4096 ? kernel
At least the next conditions should be fulfilled to hit this issue: ++> objects defined in BRIDGE-MIB are being polled; ++> polling rate should be really high - multiple requests per second; ++> number of vlan/mst*port instances should be high as well - in this certain case it was ~750 ports per two MST regions
++> decrease polling rate of BRIDGE-MIB objects
None
Click on a version to see all relevant bugs
Cisco Integration
Learn more about where this data comes from
Bug Scrub Advisor
Streamline upgrades with automated vendor bug scrubs
BugZero Enterprise
Wish you caught this bug sooner? Get proactive today.
