Vendor560 problems· 4 reviewed

Cisco

560 incident problems in Cisco environments.

All problems (560)

NETWORK-111Consistent hashing is disabled during a line-card replacement and long-lived sessions reset across ECMP membersRedundancy remains, but user sessions drop because path stickiness disappeared during the hardware swap window.NetworkAdvanced17 minPro NETWORK-122eBGP TTL security is configured on one side only after the provider path gains an extra hop, and the session drops permanentlyBoth peers still have the right neighbors configured, yet the control-plane hardening assumption no longer matches reality.NetworkAdvanced17 minPro NETWORK-1207Native VLAN assumptions hide the real issueA multi-vendor trunk stays healthy on paper, but one service VLAN never crosses it.NetworkIntermediate17 minPro NETWORK-1274OSPF neighbors form but one network never appearsAn operator proves OSPF is up and still cannot explain why only one important network is absent downstream.NetworkAdvanced17 minPro NETWORK-137VRRP advertisements pass, but the mixed-vendor pair uses different packet version expectations and both routers think they are masterLayer-3 reachability exists, yet first-hop symmetry collapses because redundancy control messages are interpreted differently.NetworkAdvanced17 minPro NETWORK-143A BGP blackhole community is set correctly, but the route reflector policy strips it before it reaches the transit edge and mitigation never activatesMitigation signaling exists at the source, yet it vanishes in the core policy path.NetworkAdvanced18 minPro NETWORK-1261A BGP policy stops preferring the intended upstreamNetwork incident scenario used for structured troubleshooting practice.NetworkAdvanced18 minPro NETWORK-1266A BGP policy stops preferring the intended upstreamAn operator refactors an inbound route-map and later discovers traffic stopped preferring the expected provider without the BGP session ever dropping.NetworkAdvanced18 minPro NETWORK-234A BGP session remains established while graceful restart keeps stale forwarding alive longer than the outage can tolerate during a failover rehearsalControl-plane continuity looks stable while data-plane truth is already different. Normal traffic masked the issue until the standby or alternate path became active under rehearsal conditions.NetworkAdvanced18 minPro NETWORK-1246A failover appears clean on the routers but users still blackholeA maintenance failover completes and a subset of endpoints remain broken until caches naturally refresh.NetworkAdvanced18 minPro NETWORK-1233A failover design works in tests but production users still blackholeA failover event completes and a subset of clients remain broken even though the new active gateway is online and forwarding.NetworkAdvanced18 minPro NETWORK-1241A failover path looks clean but one segment still blackholesA maintenance failover succeeds on the routers, but a subset of users remain broken until cache state ages out or is refreshed.NetworkAdvanced18 minPro NETWORK-204A failover signal tracks device liveness while the service behind it is already gone during a failover rehearsalThe route follows a reachable box instead of the actual service dependency. Normal traffic masked the issue until the standby or alternate path became active under rehearsal conditions.NetworkAdvanced18 minPro NETWORK-1285A firewall cluster syncs configuration but not connection state, so asymmetric return traffic after failover only breaks one long-lived application protocolPolicies match and sessions establish initially, yet one application fails after failover because session state was not replicated the way operators assumed.NetworkAdvanced18 minPro NETWORK-1280A firewall migration preserves ACL intent but one application still failsA firewall migration keeps most services working and leaves one application path failing in a way the old ACL logic did not predict.NetworkAdvanced18 minPro NETWORK-336A gateway IP remains stable while one downstream security policy still keys off the previous virtual MAC after a fabric migration during a failover rehearsalAddress continuity masks an identity change that another control still cares about. Normal traffic masked the issue until the standby or alternate path became active under rehearsal conditions.NetworkAdvanced18 minPro NETWORK-258A GRE keepalive stays up while recursive routing toward the tunnel destination loops through the tunnel itself during a failover rehearsalThe health signal survives on a path definition that undermines the tunnel's own reachability. Normal traffic masked the issue until the standby or alternate path became active under rehearsal conditions.NetworkAdvanced18 minPro NETWORK-294A GRE over IPsec path stays established (Rollout Stuck)A GRE over IPsec path stays established (Rollout Stuck) focuses on wan-and-overlay-connectivity and asks the reader to isolate Rollout Stuck in Cisco. 실무에서는 rollout-stuck 이슈를 볼 때 DNS, 라우팅, ACL/방화벽, 목적지 응답을 계층별로 잘라서 확인하면 장애 구간을 훨씬 빠르게 좁힐 수 있습니다.NetworkAdvanced18 minPro NETWORK-176A mitigation community is set at the source and stripped before it reaches the edge that should act on it after a control-plane upgradeThe intent exists in one domain and disappears in the middle of the network. The workload or policy had been stable, but the upgraded control layer now interprets one dependency differently.NetworkAdvanced18 minPro NETWORK-173A mitigation community is set at the source and stripped before it reaches the edge that should act on it after a maintenance template changeThe intent exists in one domain and disappears in the middle of the network. The path looked stable before the template changed, but one inherited assumption no longer matches the live environment.NetworkAdvanced18 minPro