Symptoms
Årsaken til den nye innlastingen vises i utdataene for kommandoen nedenfor etter at du har lastet inn på nytt:
show system reset-reason
Dette problemet er spesifikt for MDS 9148S- og MDS 9250i-plattformer. Det påvirker MDS-versjonene 8.1 (x) til og med 8.3 (x) utelukkende.
show system reset-reason
----- reset reason for module 1 (from Supervisor in slot 1) ---
1) At 374295 usecs after Sun Aug 6 19:05:46 2023
Reason: Kernel Panic
Service:
Version: 8.3(2)
2) At 554068 usecs after Wed May 10 01:10:25 2023
Reason: Kernel Panic
Service:
Version: 8.3(2)
cpp_di_si=0vip_cpu_srvc_init_kspace: Called**** TOTAL PORTS = 48 *****[sched_delayed] sched: RT throttling activatedsock: process `snmpd' is using obsolete setsockopt SO_BSDCOMPATvip_cpu_srvc_init_kspace: Calledvip_cpu_srvc_trigger_kspace: Calledin viperk_cpu_mts_init *mts_q:0xd46642c0 mts_q :0xd3989bd8During ISSU MTS init handler for Cpu is loadedmts_q ::0xd46642c0 vip_cpu_srvc_trigger_kspace: Calledsmhb_mod_hb_params: Sysmgr modifying HB params, hb_intvl 2 max_hb_loss 8smhb_enable_disable_wd: do nothing on lc_on_hybrid_supps (10188) used greatest stack depth: 4432 bytes leftps (11005) used greatest stack depth: 4208 bytes leftlibphy: mdio@fff726520:00 - Link is Downlibphy: mdio@fff726520:00 - Link is Up - 1000/FullKernel stack overflow in process dbfdc9b0, r1=d64a9afcKernel panic - not syncing: kernel stack overflowKGDB: Waiting for remote debuggerStart stack dumpingMoving to kernel stackDone stack dumpingStart register dumpingDone register dumpingDone all dumping 4053 8196
**** KERNEL PANIC OCCURED*******Writing reset reason.
Irqs 1
Writing stack trace
Writing kernel traces
Starting dump of trace eventsUnable to handle kernel paging request for data at address 0x9596a008
Faultiting instruction address: 0xc00b9804`
2023 Aug 6 21:03:30 SGMCISWXXXX1 %SYSMGR-5-MODULE_ONLINE: System Manager has received notification of local module becoming online.
show logging log
2023 Aug 6 21:02:52 SGMCISWXXXX1 %SYSLOG-2-SYSTEM_MSG : Syslogs wont be logged into logflash until logflash is online
2023 Aug 6 21:02:53 SGMCISWXXXX1 %KERN-5-SYSTEM_MSG: [ 0.280269] SCSI subsystem initialized - kernel
2023 Aug 6 21:02:53 SGMCISWXXXX1 %KERN-4-SYSTEM_MSG: [ 0.348776] pci 0001:03:00.2: EHCI: unrecognized capability 00 - kernel
2023 Aug 6 21:02:53 SGMCISWXXXX1 %KERN-4-SYSTEM_MSG: [ 0.409937] bounce pool size: 64 pages - kernel
2023 Aug 6 21:02:53 SGMCISWXXXX1 %KERN-4-SYSTEM_MSG: [ 0.689291] kworker/u4:0 (824) used greatest stack depth: 6400 bytes left - kernel
2023 Aug 6 21:02:53 SGMCISWXXXX1 %KERN-5-SYSTEM_MSG: [ 0.804272] physmap platform flash device: 02000000 at f0000000 - kernel
2023 Aug 6 21:02:53 SGMCISWXXXX1 %KERN-3-SYSTEM_MSG: [ 0.804302] physmap-flash physmap-flash.0: Could not reserve memory region - kernel
2023 Aug 6 21:02:53 SGMCISWXXXX1 %KERN-4-SYSTEM_MSG: [ 0.886657] physmap-flash: probe of physmap-flash.0 failed with error -12 - kernel
2023 Aug 6 21:02:53 SGMCISWXXXX1 %KERN-4-SYSTEM_MSG: [ 1.052036] mpc85xx_mc_err_probe: No ECC DIMMs discovered - kernel
2023 Aug 6 21:02:53 SGMCISWXXXX1 %KERN-0-SYSTEM_MSG: [ 1.074666] Enabling all PCI devices - kernel
2023 Aug 6 21:02:53 SGMCISWXXXX1 %KERN-3-SYSTEM_MSG: [ 16.171032] CMOS: Module initialized - kernel
2023 Aug 6 21:02:53 SGMCISWXXXX1 %KERN-4-SYSTEM_MSG: [ 46.307515] ICMPv6: process `sysctl' is using deprecated sysctl (syscall) net.ipv6.neigh.default.base_reachable_time - use net.ipv6.neigh.default.base_reachable_time_ms instead - kernel
2023 Aug 6 21:02:53 SGMCISWXXXX1 %KERN-4-SYSTEM_MSG: [ 46.335132] nr_pdflush_threads exported in /proc is scheduled for removal - kernel
2023 Aug 6 21:02:53 SGMCISWXXXX1 %KERN-4-SYSTEM_MSG: [ 46.336491] sysctl: The scan_unevictable_pages sysctl/node-interface has been disabled for lack of a legitimate use case. If you have one, please send an email to linux-mm@kvack.org. - kernel
2023 Aug 6 21:03:24 SGMCISWXXXX1 %FCS-5-API_FAIL: %$VSAN 120%$ pm_send_get_ports_in_vsans() failed: fu ha standby message queued
2023 Aug 6 21:03:24 SGMCISWXXXX1 %FCS-5-API_FAIL: %$VSAN 150%$ pm_send_get_ports_in_vsans() failed: fu ha standby message queued
2023 Aug 6 21:03:27 SGMCISWXXXX1 %FCDOMAIN-5-DOMAIN_TYPE_IS_PREFERRED: The domain ID type is currently configured as preferred in all the existing VSANs
2023 Aug 6 21:03:28 SGMCISWXXXX1 %DAEMON-3-SYSTEM_MSG: sendto(10.XX.X.48): Network is unreachable - ntpd[3589]
2023 Aug 6 21:03:28 SGMCISWXXXX1 %DAEMON-3-SYSTEM_MSG: sendto(10.XX.X.49): Network is unreachable - ntpd[3589]
2023 Aug 6 21:03:30 SGMCISWXXXX1 %MODULE-5-ACTIVE_SUP_OK: Supervisor 1 is active (Serial number: JAE200XXXXV)
2023 Aug 6 21:03:30 SGMCISWXXXX1 %PLATFORM-5-MOD_STATUS: Module 1 current-status is MOD_STATUS_ONLINE/OK
Cause
Sysmgr-tjenestekrasj forekommer på MDS 9148S- og MDS 9250i-plattformer som kjører MDS-utgivelser 8.1 (x) til og med 8.3 (x). Det fører til kjernekrise, noe som fører til at hele svitsjen ikke er tilgjengelig. Resultatet er at systemet laster inn på nytt for å gjenopprette funksjonaliteten. På grunn av kjernekrise er det ikke mulig å fastslå den nøyaktige årsaken bak avslutningen av sysmgr-tjenesten.
Resolution
Løsningen:
Det finnes ingen kjent løsning for å løse dette problemet. Last imidlertid inn svitsjen på nytt for å frigjøre den utilgjengelige tilstanden.
Oppløsning:
Forslaget fra TAC er å oppgradere koden til versjon 8.4 (2f).
Kjente berørte utgivelser: 8.3(2)
Cisco-problem-ID: CSCvu16450 og
CSCvp13486
Cisco TAC-sak: 695961883
Affected Products
Connectrix MDS 9148S