Launching soon: The world's first vendor agnostic bug scrubLearn more & join waitlist
OPERATIONAL DEFECT DATABASE
...
...
Ingress packets being dropped on NP counter PARSE_DROP_IN_UIDB_TCAM_MISS . Also in np drvlog file there is: Nov 1 02:02:49.485 ERROR! 0x80001755 EZprmCAMC_ExtTCAMCommand: command fails due to EZprmCAMC_TCAM_CONTROL_INTERFACE_OR_DEVICE_ERR in file 'drivers/chips/np/ezchip-5c/src/host/driver/src/prm/chn/EZprmCAMC.c' line 2493 Nov 1 02:02:49.485 ERROR! 0x80001755 EZprmCAMC_ExtTCAMCommand: command fails due to TCAM error in file 'drivers/chips/np/ezchip-5c/src/host/driver/src/prm/chn/EZprmCAMC.c' line 2499 Nov 1 02:02:49.485 ERROR! 0x80001755 EZprmCAMC_ExtTCAMCommand: Stored error. in file 'drivers/chips/np/ezchip-5c/src/host/driver/src/prm/chn/EZprmCAMC.c' line 2581 Nov 1 02:02:49.486 ERROR! 0x80001755 EZapiPrm_ExtTCAMCommand: EZprmCAMC_ExtTCAMCommand failed, channel Id 0. in file 'drivers/chips/np/ezchip-5c/src/host/driver/src/api/EZapiPrm.c' line 1016 drvlog file can be copied/seen on disk0:np for affected card or displayed with 'show controllers np drvlog location {}' Another symptom: BFD may be down as a consequence of LACP being down for those (and all) bundles: #show bfd all session Interface Dest Addr Local det time(int*mult) State Echo Async H/W NPU ------------------- --------------- ---------------- ---------------- ---------- BE4.905 xx.xx.xx.xx 0s 6s DOWN #show bundle status detail Bundle-Ether2 (Not operational): WARNING: Not enough selectable links to be operational TenGigE0/0/0/14 (Not active): WARNING: Link is Defaulted; LACPDUs are not being received from the partner
Normal operation. Issue is very rarely seen. The above described TCAM errors are present in the NPdrvlog file, but there are no TCAM interrupts seen: RP/0/RSP0/CPU0:PRM5#show controllers np interrupts all all loc 0/3/cpu0 | include TCAM Fri May 15 09:27:19.693 EDT 0 TCAM 30 0 0 N Y 0x561e35784861 <---- this shows 0 interrupts seen Here is an example with 4 interrupts seen: RP/0/RSP0/CPU0:PRM5#show controllers np interrupts all all loc 0/2/cpu0 | include TCAM Fri May 15 09:26:59.021 EDT 0 TCAM 30 4 0 Y Y 0x556f48417861 <---- this shows 4 interrupts seen
No workaround but please try a LC reload to recover - it is also possible that a manual NP Fast Reset could help with the TCAM Error interrupt not being seen
This is enhancement to incorporate TCAM Error polling and automatic recovery mechanism for such issues. Issue is not SW related and also it can affect any linecard type from Typhoon/Tomahawk families. Please note that the TCAM error detection and correction processing is primarily focused on addressing Parity Errors seen in the TCAM; however, the 32-bit (cXR) version of the TCAM driver does not currently support Parity Error detection / correction. So while the TCAM Error polling code will potentially detect TCAM NP interface errors on the 32-bit images used for Typhoon & Tomahawk family LCs, the TCAM Error polling code has not been verified to detect and correct the TCAM NP interface errors using 32-bit images. Also note that the TCAM Error polling enhancement added for this bug was only done for Tomahawk code.