Network Protocol Timer Reference
Default timers, retransmit counters, and dead intervals for common L2/L3 protocols and infrastructure services. Values reflect vendor and RFC defaults — always verify against your platform's documentation.
⌕
◎
No timers match — try a different search term or category.
Layer 2 — Data Link Spanning Tree · Discovery · Link Aggregation · ARP
STP / RSTP / MSTP Spanning Tree Protocol — IEEE 802.1Q (incorporates 802.1D, 802.1w, 802.1s) IEEE 802.1Q-2022 ↗
Timer Default Unit Notes
Hello Time 2 sec Interval between configuration BPDUs transmitted by the root bridge (STP and RSTP).
Forward Delay 15 sec Time spent in Listening and Learning states (STP). Used for backward compatibility timers in RSTP.
Max Age 20 sec Maximum BPDU age. A bridge discards a stored BPDU and triggers topology change when this is exceeded.
Topology Change Timer 35 sec STP only: duration of TCN flooding = Forward Delay + Max Age (15 + 20). Not a configurable timer — derived value.
RSTP Migration Delay 3 sec Delay before an RSTP port returns to RSTP mode after receiving legacy STP BPDUs. Prevents premature migration.
MSTP Max Hops 20 count Maximum number of hops an MST BPDU can traverse within a region before being discarded. MSTP (802.1s) only.
CDP Cisco Discovery Protocol — Cisco Proprietary Cisco Docs ↗
Timer Default Unit Notes
Send Timer (Hello) 60 sec Interval between CDP advertisement (hello) transmissions on each enabled interface.
Holdtime 180 sec Time a neighbor entry is retained after the last CDP packet was received. Defaults to 3× send timer.
LLDP Link Layer Discovery Protocol — IEEE 802.1AB IEEE 802.1AB-2016 ↗
Timer Default Unit Notes
TX Interval 30 sec Interval between regular LLDP frame transmissions on each enabled port.
Hold Multiplier 4 count TTL advertised in LLDP TLV = TX Interval × Hold Multiplier. Default TTL = 120 sec. Neighbor removed when TTL expires.
Reinitialization Delay 2 sec Minimum time before LLDP re-initializes on a port that was disabled. Prevents rapid restart loops.
TX Delay 2 sec Minimum delay between successive LLDP transmissions triggered by a local change. Must be ≤ 0.25 × TX Interval.
Fast Start TX Period 1 sec Rapid transmission interval when a port first becomes active (fast-start). Sends fast-start count frames before resuming normal TX interval.
Fast Start Count 4 count Number of rapid LLDP frames transmitted at fast-start TX period when a port comes up.
LACP / PAgP Link Aggregation Control Protocol / Port Aggregation Protocol — IEEE 802.1AX IEEE 802.1AX-2020 ↗
Timer Default Unit Notes
LACP Fast TX Rate 1 sec LACPDU transmission interval in fast (short-timer) mode. Configured per-port on Cisco with lacp rate fast.
LACP Slow TX Rate 30 sec LACPDU transmission interval in normal (slow-timer) mode. Default on most platforms.
LACP Fast Timeout 3 sec Partner declared dead after 3 missed fast-rate LACPDUs (3 × 1 sec). Enables sub-second failover with fast rate on both sides.
LACP Slow Timeout 90 sec Partner declared dead after 3 missed slow-rate LACPDUs (3 × 30 sec). Standard fallback when fast timers are not configured.
PAgP Hello (Cisco) 30 sec Interval between PAgP hello packets. Cisco proprietary; no published RFC. Not interoperable with LACP.
PAgP Holdtime (Cisco) 90 sec Time before a PAgP peer is declared dead if no hellos are received (3 × hello interval).
ARP Address Resolution Protocol — RFC 826 RFC 826 ↗
Timer Default Unit Notes
Cache Timeout (Cisco IOS) 240 min ARP entry lifetime in the Cisco IOS ARP cache. Set with arp timeout per interface (default 4 hours = 14400 sec).
REACHABLE State (Linux) 30 sec Time a Linux neighbor entry stays in REACHABLE state after confirmation. Randomized: base_reachable_time / 2 to 3 × base_reachable_time / 2 (default base = 30 sec).
gc_stale_time (Linux) 60 sec Time before a STALE neighbor entry triggers probe (DELAY → PROBE). After failed probes, entry moves to FAILED state.
Cache Timeout (Windows) 600 sec Default dynamic ARP entry lifetime on Windows. Configurable; range typically 15 sec – 10 min depending on OS version.
Request Retries 3 count Number of ARP request retransmissions before declaring a host unreachable. Implementation-specific; RFC 826 does not define a retry count.
Layer 3 — Routing Protocols OSPF · EIGRP · BGP · RIP · IS-IS
OSPF Open Shortest Path First (OSPFv2 / OSPFv3) — RFC 2328 / RFC 5340 RFC 2328 ↗
Timer Default Unit Notes
Hello — Broadcast / P2P 10 sec Hello interval on broadcast (Ethernet) and point-to-point interfaces. Must match on both sides to form adjacency.
Hello — NBMA / P2MP 30 sec Hello interval on non-broadcast multi-access and point-to-multipoint interfaces (e.g., Frame Relay).
Dead Interval — Broadcast / P2P 40 sec Neighbor declared dead after no Hello received in this interval. Defaults to 4× Hello. Must match on both sides.
Dead Interval — NBMA / P2MP 120 sec Dead interval for NBMA and P2MP networks (4× 30-sec hello).
Retransmit Interval 5 sec Time to wait before retransmitting an unacknowledged LSA to a neighbor. Increase on high-latency links.
Transmit Delay 1 sec Estimated time to transmit a link-state update. Added to LSA age before transmission. Rarely changed in practice.
Wait Interval 40 sec Time a router waits to trigger DR/BDR election on startup if no Hello is received. Equal to Dead Interval.
LSA Refresh Time 1800 sec How often an originating router refreshes its own LSAs before they age out. Equals half of LSA Max Age (30 min).
LSA Max Age 3600 sec Maximum age of an LSA. An LSA reaching Max Age is flushed from the LSDB and flooded with age = MaxAge (1 hour).
SPF Initial Delay 0–200 ms Delay before running SPF after a topology change. Modern platforms use SPF throttle (e.g., Cisco IOS XE default: 200 ms initial, 1 sec hold, 10 sec max).
EIGRP Enhanced Interior Gateway Routing Protocol — RFC 7868 RFC 7868 ↗
Timer Default Unit Notes
Hello — LAN / High Bandwidth 5 sec Hello interval on interfaces with bandwidth > T1 (1.544 Mbps), including Ethernet and fast interfaces.
Hello — WAN / Low Bandwidth 60 sec Hello interval on multipoint or interfaces with bandwidth ≤ T1. Also used on point-to-multipoint WAN topologies.
Hold Time — LAN 15 sec Neighbor declared unreachable if no Hello received within this interval (3× hello). Advertised in Hello packets.
Hold Time — WAN 180 sec Neighbor dead timer on low-bandwidth WAN interfaces (3× 60-sec hello).
Active Timer (SIA) 3 min Maximum time to wait for a feasible successor reply during active state. A neighbor that fails to reply is declared stuck-in-active (SIA) and reset. Default: 180 sec.
Retransmission Timeout (RTO) dynamic ms Per-neighbor; computed from Smooth Round-Trip Time (SRTT). Minimum ~200 ms, maximum 5000 ms. Doubles on each retransmit up to 16 retries.
Max Retransmissions 16 count Maximum number of RTO-based retransmissions before a neighbor is declared unreachable and the adjacency is reset.
BGP Border Gateway Protocol v4 — RFC 4271 RFC 4271 ↗
Timer Default Unit Notes
Hold Time 180 sec Session terminated if no KEEPALIVE or UPDATE received. Negotiated to the lower of the two peers' values. RFC minimum is 90 sec; 0 disables keepalives.
Keepalive Interval 60 sec Interval between KEEPALIVE messages. Defaults to ⅓ of negotiated Hold Time. Must be received before Hold Time expires to maintain session.
Connect Retry 120 sec Time to wait before retrying a TCP connection to a peer after a failure. Applies in Idle/Connect/Active states per FSM.
Min Route Adv. Interval — eBGP 30 sec Minimum interval between successive UPDATE advertisements to eBGP peers (MRAI). Reduces UPDATE churn during route instability.
Min Route Adv. Interval — iBGP 5 sec MRAI for iBGP peers. Lower default reflects trust between internal peers and desire for faster convergence within an AS.
Graceful Restart Stale Time 360 sec Maximum time to retain stale routes from a restarting peer (RFC 4724). Routes are marked stale and retained during the restart window.
Idle Hold Time 30 sec Time BGP remains in Idle state after a session reset before attempting reconnect. Cisco IOS default; increases with repeated failures (up to 64 sec).
RIPv2 Routing Information Protocol v2 — RFC 2453 RFC 2453 ↗
Timer Default Unit Notes
Update Timer 30 sec Interval at which RIP sends full routing table updates to neighbors. Jittered ±5 sec to prevent synchronization.
Invalid Timer 180 sec Route marked invalid (metric = 16 / unreachable) if no update is received for this duration. Resets on each received update (6× update timer).
Holddown Timer 180 sec Duration a poisoned route is suppressed to prevent routing loops after invalidation. Only updates with a better metric can override during holddown.
Flush Timer 240 sec Route completely removed from routing table if not refreshed within this time from the last valid update (8× update timer).
IS-IS Intermediate System to Intermediate System — RFC 5308 / ISO 10589 RFC 5308 ↗
Timer Default Unit Notes
Hello Interval — LAN (L1/L2) 10 sec IIH (IS-IS Hello) PDU transmission interval on broadcast (LAN) interfaces for Level-1 and Level-2.
Hello Multiplier — LAN 3 count Holding time advertised = Hello × Multiplier. Neighbor declared dead after Hold Time with no IIH received. Default: 30 sec hold.
Hello Interval — P2P 10 sec IIH transmission interval on point-to-point interfaces.
Hello Multiplier — P2P 3 count P2P hold time = Hello × Multiplier (default: 30 sec). Adjacency reset if Hold Time expires without receiving an IIH.
LSP Gen Interval 30 sec Minimum interval between generating new LSPs originating from this router. Throttles flooding during instability. Platform-specific.
LSP Refresh Interval 900 sec How often the originating IS refreshes its LSPs before they expire. Must be less than Max LSP Lifetime (default: 15 min).
Max LSP Lifetime 1200 sec Maximum age of an LSP in the LSDB before it is purged. Must be greater than LSP Refresh Interval (default: 20 min).
CSNP Interval 10 sec Interval between Complete Sequence Number PDU transmissions on broadcast interfaces. DIS-only function; used to synchronize LSDBs.
Retransmit Interval (P2P) 5 sec Time before retransmitting an unacknowledged LSP or PSNP on point-to-point links.
Gateway Redundancy HSRP · VRRP · GLBP
HSRP Hot Standby Router Protocol v1 / v2 — RFC 2281 (v1) RFC 2281 ↗
Timer Default Unit Notes
Hello Interval 3 sec Interval between HSRP Hello messages. HSRPv2 supports millisecond timers (minimum 15 ms). Configure with standby X timers.
Hold Time 10 sec Active router declared dead if no Hello received within this interval. Should be ≥ 3× Hello. Must be consistent across group members.
Preempt Delay (Reload) 0 sec No built-in preempt delay by default. Best practice: configure a reload delay (60–180 sec) so routing protocols converge before preemption occurs.
VRRP Virtual Router Redundancy Protocol v3 — RFC 5798 RFC 5798 ↗
Timer Default Unit Notes
Advertisement Interval 1 sec Interval between VRRP Advertisement messages sent by the Master. VRRPv3 supports centisecond granularity.
Master Down Interval ~3.6 sec Time a Backup waits before taking over. Formula: (3 × Adv_Interval) + Skew_Time. Skew = (256 − Priority) / 256. With default priority 100: ≈ 3.609 sec.
Preempt Delay 0 sec Additional delay before a higher-priority backup preempts the current master. Best practice: configure 30–60 sec to allow routing convergence.
GLBP Gateway Load Balancing Protocol — Cisco Proprietary Cisco Docs ↗
Timer Default Unit Notes
Hello Interval 3 sec Interval between GLBP Hello messages among group members.
Hold Time 10 sec Time before declaring a GLBP peer (AVG or AVF) inactive if no Hello is received. Must be ≥ 3× Hello.
Redirect Timer 600 sec After an AVF (Active Virtual Forwarder) fails, the AVG redirects ARP to another forwarder for this duration (10 min) before stopping.
Forwarder Timeout 14400 sec Time before a failed forwarder's virtual MAC address is flushed from the network. Hosts must have updated ARP before this expires (4 hours).
Infrastructure Services DHCP · DNS · NTP · RADIUS · 802.1X
DHCP Dynamic Host Configuration Protocol (DHCPv4) — RFC 2131 RFC 2131 ↗
Timer Default Unit Notes
Initial Discover Delay 0–10 sec Random delay before the first DHCPDISCOVER to desynchronize clients. RFC 2131 recommendation; implementation-specific.
Retransmit Backoff 4 / 8 / 16 / 32 sec Exponential backoff between DHCPDISCOVER retransmissions. Each interval doubles; capped at 64 sec maximum. ±1 sec randomization applied per RFC.
Max Discover Retries 4 count Number of DHCPDISCOVER retransmissions before the client gives up. Implementation-specific (RFC 2131 does not mandate a fixed count).
Lease Renewal (T1) 50 % At 50% of lease time, client unicasts DHCPREQUEST to the server to renew. With a 24-hour lease: T1 = 43200 sec (12 hours).
Lease Rebind (T2) 87.5 % At 87.5% of lease time, client broadcasts DHCPREQUEST to any available server. With a 24-hour lease: T2 = 75600 sec (21 hours).
Default Lease Time 86400 sec Common server default lease duration (24 hours). Varies by implementation; server-operator configurable. RFC 2131 does not specify a default.
DNS Domain Name System — RFC 1034 / RFC 1035 RFC 1035 ↗
Timer Default Unit Notes
Query Timeout (Cisco IOS) 3 sec Time Cisco IOS waits for a DNS response before retrying or trying the next configured name server.
Query Timeout (resolver) 5 sec Typical OS resolver query timeout per attempt. Defined in /etc/resolv.conf (Linux); options timeout:5. RFC 1035 leaves this configurable.
Retry Count 2–4 count Retries per name server before trying the next. Cisco IOS: 2; Linux resolv.conf default: 2; Windows: 3–4. RFC 1035 does not mandate a value.
Negative Cache TTL 300 sec How long NXDOMAIN (non-existent domain) responses are cached. Sourced from the SOA record's minimum field (RFC 2308). Typical operational range: 300–900 sec.
Positive TTL (common) 300–3600 sec Zone-operator-configured TTL for positive records. 300 sec (5 min) is common for dynamic or frequently changing records; 3600 sec (1 hr) for stable records.
NTP Network Time Protocol v4 — RFC 5905 RFC 5905 ↗
Timer Default Unit Notes
Min Poll Interval 64 sec Minimum polling interval (2⁶ seconds). Used when the clock is unsynchronized, offset is large, or a new association is starting. RFC 5905 default: minpoll = 6 (2⁶).
Max Poll Interval 1024 sec Maximum polling interval (2¹⁰ seconds ≈ 17 min). Used when the clock is stable. RFC 5905 default: maxpoll = 10. Cisco IOS default maxpoll = 10 (1024 sec).
Reachability Shift Register 8 count 8-bit shift register tracking last 8 polls. Server declared unreachable if all bits are 0 (8 consecutive failures). RFC 5905 §11.4.
Step Threshold 128 ms Clock offset above which ntpd applies a step correction (instantaneous jump) rather than gradual slew. RFC 5905 default. Configurable with tinker stepout.
Panic Threshold 1000 sec Offset above which ntpd refuses to correct the clock and exits (protects against erroneous time jumps). Disabled with tinker panic 0.
Clock Slew Rate 500 ms Maximum rate at which ntpd gradually adjusts the clock (500 ppm = 500 ms/s). RFC 5905 §11.2. Very large offsets below panic threshold may take hours to converge.
RADIUS Remote Authentication Dial-In User Service — RFC 2865 RFC 2865 ↗
Timer Default Unit Notes
Request Timeout (Cisco IOS) 5 sec Time to wait for a RADIUS Access-Accept/Reject/Challenge response before retransmitting or trying the next server. Configured globally or per server.
Max Retransmits 3 count Number of times a RADIUS request is retransmitted before the server is considered unresponsive. Total attempts = 1 + retransmits = 4.
Deadtime 0 min When > 0, time an unresponsive server is skipped before re-attempting. Default 0 = deadtime disabled; server is always tried. Best practice: set to 5–15 min.
Dead Criteria — Time 10 sec Minimum time window for evaluating RADIUS failures before marking a server dead (Cisco IOS dead criteria detection). Works with Tries counter.
Dead Criteria — Tries 10 count Minimum failed transactions within the Dead Criteria Time window before the server is automatically marked dead.
Session Timeout (802.1X) 3600 sec RADIUS Session-Timeout attribute (Attribute 27) supplied to the NAS for 802.1X reauthentication scheduling. Server-configured; overrides local reauth-period when sent.
802.1X Port-Based Network Access Control — IEEE 802.1X-2020 IEEE 802.1X-2020 ↗
Timer Default Unit Notes
TX Period 30 sec Time between EAP-Request/Identity retransmissions to a supplicant that has not responded. Cisco default; IEEE 802.1X standard names this txPeriod.
Supplicant Timeout 30 sec Time the authenticator waits for a response to an EAP message sent to the supplicant before retransmitting. IEEE 802.1X: suppTimeout.
Server Timeout 30 sec Time the authenticator waits for a response from the RADIUS server. If exceeded, authentication fails and the port enters a failure state. IEEE 802.1X: serverTimeout.
Quiet Period 60 sec Hold-off period after a failed authentication during which the authenticator will not initiate a new authentication attempt. IEEE 802.1X: quietPeriod.
Reauthentication Period 3600 sec Interval between periodic reauthentications for an authenticated port (1 hour). Can be overridden by RADIUS Session-Timeout attribute from the server.
Max Requests 2 count Maximum EAP retransmissions to the supplicant during authentication. After max-requests, authentication fails and quiet period begins.
Max Reauth Requests 2 count Maximum retransmissions during the reauthentication phase. Separate from initial auth max-requests.