Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

Resolved Issues

 

This section lists the issues fixed in the Junos OS main release and the maintenance releases. The identifier following the description is the tracking number in the Juniper Networks Problem Report (PR) tracking system.

Resolved Issues: 15.1R7

General Routing

  • After the primary clock goes down, it locks to the previous secondary as the new primary. The clock should lock to secondary. PR1114281

  • On PTX Series routers, a faulty power supply module (PSM) might generate excessive interrupt requests. These hardware interrupt requests are processed by the chassisd process and might restart the chassisd process when the condition persists for more than 200 seconds. PR1226992

  • When a user configures a TPID value other than 0x8100 on a single-tagged interface with the configuration command vlan-tags outer TPID-VLAN-ID, the TPID value 0x8100 is used instead of the user-specified TPID value. PR1237687

  • The kernel log message mastership: sent other RE mastership loss signal might be displayed frequently during normal operation on a PTX5000 backup Routing Engine. The message is cosmetic and does not indicate any service impact or Routing Engine mastership loss. PR1260884

  • 100Base-ER4 (740-045420) is shown as UNKNOWN in the output of the show chassis hardware command in Junos OS Release 15.1R5.5. This is a non-service-impacting regression. PR1280089

  • In a PTX3000 with SIB-SFF-PTX-240-S, after the device is rebooted, FPCs might go offline because of fabric healing. This is a PTX3000 Switch Interface Board (SIB) issue. When the issue occurs, the system restarts the FPC. There is no impact to the existing working FPC. The issue can be resolved by taking the SIBs offline one by one without impacting the traffic throughput. PR1282983

  • On PTX Series routers, when a multicast route and a next hop for it are being programmed into the Packet Forwarding Engine, part of the memory allocated for the next-hop structure might get lost, resulting in heap memory leak on the FPC. PR1302303

  • The hold timer enables interface damping by not advertising interface transitions until the hold timer duration is over. On PTX5000 with optical interfaces, if the interface hold-time down timer is configured to be less than 500 nbsp ms, the timer does not always work. PR1307302

Infrastructure

  • The show system users command output displays users that are not using the router. The request system logout CLI command cannot clear the stale Telnet sessions. This is a cosmetic issue, because the show system connection command and the CLI process show only the current session. user@router> show system users 5:39PM up 8 mins, 3 users, load averages: 0.27, 0.43, 0.26 USER TTY FROM LOGIN@ IDLE WHAT lab pts/0 172.27.208.216 5:36PM - -cli (cli) <---- old telnet session lab pts/0 172.27.208.216 5:38PM - -cli (cli) <---- old telnet session lab pts/0 172.27.208.216 5:39PM - -cli (cli) <---- current telnet session user@router> show system connections |match 172.27.208.216 tcp4 0 0 172.27.116.36.23 172.27.208.216.63830 ESTABLISHED user@router> start shell % ps -aux |grep cli|grep -v grep lab 21016 0.0 0.2 786268 50304 0 S 5:39PM 0:00.15 -cli (cli) %. PR1247546

Interfaces and Chassis

  • When Routing Engine switchover is executed, the dcd process performs a check on the aggregated Ethernet interface. The check fails if the aggregated Ethernet interface has a member interface with framing settings. The failed check triggers flapping of both the aggregated Ethernet interface and its member interface. PR1287547

MPLS

  • In an MPLS OAM environment, a rare timing condition can result in rpd process crash when a memory clean task is delayed. PR1233042

  • In an MPLS environment, when a non-master routing instance with label-switched paths (LSPs) is deleted and re-added, the rpd process might crash. As a result, the routing protocols are impacted and traffic disruption due to loss of routing information is observed. PR1241631

  • If there is an error during the creation of the RSVP Path state (the PSB data structure), the data structure itself is freed but some associated memory is not freed. This situation causes a memory leak. It is very unlikely that this error condition ever happens on an NSR master Routing Engine (or when no NSR is configured). But on the NSR backup Routing Engine, there are more likely to be conditions that cause the Path state creation to fail, thus exposing the memory leak in the error-handling code. Thus this memory leak seen on the NSR backup Routing Engine. The fix went to address mitigation of memory leak due to RSVP_HOP object in this PR. PR1328974

Network Management and Monitoring

  • If max-events-queued is not configured, then the eventd process might crash when generating a large amount of logging messages. PR1155756

Platform and Infrastructure

  • An upper bound check has been introduced to avoid the condition in which the packet transmission might go into an infinite loop if the maximum number of retries is exceeded in case the transmission times out. PR1315682

Routing Protocols

  • In the rare scenario with a maximum number of routes in the BGP RIB_OUT table (for example, if there are more than 700,000 BGP routes in the route table), if link flapping is experienced on the BGP protocol, it might cause the rpd process to crash. PR1222554

  • On all platforms, if MPLS goes down because of link flapping,, FPC reboot, or FPC restart, then rpd core files are generated. PR1228388

  • When Junos interworks with other vendors device, the primary path of MPLS LSP might switch to other address even though strict is configured for primary path. PR1316861

Resolved Issues: 15.1R6

Class of Service (CoS)

  • The following error log message might be seen with hierarchical CoS and strict-high scheduling configured. Dec 27 11:08:02.293 mand-re0 fpc1 cos_check_temporal_buffer_status: IFD ge-1/2/1 IFL 358: Delay buffer computation incorrect.^M . If hierarchical scheduler is configured for a physical interface and if guaranteed rate is not set for a logical interface under this physical interface, then the temporal buffer is configured. The display of error message is valid when guaranteed rate is 0, but it is not valid when guaranteed rate is disabled. PR1238719

General Routing

  • After primary clock goes down, clock locks to previous secondary as new primary. The clock should lock to secondary. PR1114281

  • On the FPC-SFF-PTX-P1-A (PTX3000), FPC-SFF-PTX-T (PTX3000), FPC-PTX-P1-A (PTX5000), and FPC2-PTX-P1A (PTX5000), packet loss might be observed in an equal-cost multipath (ECMP) or aggregated Ethernet scenario. It occurs in a race condition: because the unilist is created before ARP has learned MAC addresses, the selector table is corrupted. PR1120370

  • In rare cases, multiple Routing Engine switchovers might result in a SNGPMB crash. (The SNG PMB is the same thing as SPMB. It is on the line card and contains the LCPU.) It also manages locally discovered issues and the switch fabric (through the chassis manager thread, which communicates with the fabric manager thread in chassisd). PR1176094

  • For PTX Series routers, the IPv6 unilist next-hop member is in "replaced" status on the Packet Forwarding Engine after interface flapping with IPv6 Neighbor Discovery timeout. While the problem is occurring, the routing table will display all right next-hop status but cannot forward traffic because forwarding next-hop in Packet Forwarding Engine is in "replaced" status and no longer active. PR1177023

  • When an ARP entry is learned through an AE interface and a route is pointing to that ARP nexthop, the ARP entry will not expire even if the ARP IP is not reachable. This issue occurs due to the route nexthop on the AE interface getting stuck in a unicast state even if the remote end is not reachable, and the RPD is unaware that the ARP is invalid. So, with this resolution, the route nexthop on the AE interface can be shown in the hold state when the remote end is not reachable. PR1211757

Infrastructure

  • In an RSVP scenario, provision RSVP label-switched path (LSP) with ldp-tunneling enabled and these LSPs are configured with link protection, continuous kernel logs, and LDP stats timeout error might be seen when executing show ldp traffic-statistics. PR1215452

Multiprotocol Label Switching (MPLS)

  • On a P2MP LSP transit router with link-protection enabled, if the LSP is the last sub-LSP, tearing the last sub-LSP (for example, a RESV tear message is received from a downstream router) might crash the routing process (rpd). PR1036452

  • When we have a statically configured ingress and transit LSPs, due to timing issue, there could be a scenario wherein the selfID used by the transit LSP gets allocated to the ingress LSP. Ingress static LSP does not reuse the same selfID across restarts whereas transit static LSP tries to reuse the same. This leads to the RPD crash from the collision when the transit LSP tried to reuse the same ID. PR1084736

  • On PTX Series platforms, the rpd might crash when the RSVP bypass undergoes re-optimization and the re-optimized instance encounters failure before it becomes the main instance. The core files can be seen by executing the CLI command show system core files. Stack trace: #0 0x0000000802ad8bd4 in patricia_node_in_tree () #0 0x0000000802ad8bd4 in patricia_node_in_tree () #1 0x00000000009ec3da in tag_pvc_shortwait () #2 0x0000000000a2fe94 in ted_delete_cc_from_link () #3 0x0000000000a3009d in ted2cspf_cleanup () #4 0x0000000000f27d56 in task_job_create_foreground () #5 0x0000000000f289e5 in task_job_bg_dispatch () #6 0x0000000000f24d85 in task_scheduler () #7 0x000000000062b9e2 in main (). PR1250253

Platform and Infrastructure

  • There is a race condition between database creation and database access. This is rarely reproducible. There is no functional impact on the core. PR1225086

VPNs

  • On PTX Series platform, the L2 circuit does not switch from primary to backup and vice versa based on the APS status change, because when APS switchover happens, the PW switchover does not switch to the new APS active neighbor. PR1239381

Resolved Issues: 15.1R5

General Routing

  • To lock to the secondary node when primary node goes down, you should not reprogram the Centralized Clock Generator (CCG). However, when you are determining whether clock_selection should be aborted, if the old primary clock source has been removed from the configuration, do not abort; new sources need to be reselected. PR1094106

  • The routing protocol process (rpd) fails to respond to any new CLI routing commands (for example, show mpls lsp terse). The rpd is forking a child process while rpd is processing a show command. When the subprocess tries to exit, it tries to close the management socket being used by the show command. This failure might cause the rpd subprocess to crash and generate a core file. It also removes the rpd pid file, which prevents the rpd from processing any new CLI commands even though original rpd process continues to run normally. PR1111526

  • On the FPC-SFF-PTX-P1-A(PTX3000), FPC-SFF-PTX-T(PTX3000), FPC-PTX-P1-A(PTX5000), and FPC2 -PTX-P1A(PTX5000), packet loss might be observed in an equal-cost multipath (ECMP) or aggregated Ethernet (AE) scenario. It occurs in a race condition: because the unilist is created before ARP has learned MAC addresses, the selector table is corrupted. PR1120370

  • On PTX series platforms with FPC3, the octets of IPv4 source and destination addresses in the firewall log are listed in reverse; this might affect troubleshooting. The IPv6 log works fine. This is a minor issue, and there is no other service impact. PR1141495

  • Because of incorrect implementation in the code, power consumption was not fetched properly for the SIBs when using PTX PDU2. PR1156265

  • FPC might crash after FPC reloading (restart FPC/non-GRES Routing Engine switchover), because of memory corruption when interface-specific filter process IPC messages. To fix this, the way firewall daemon (dfwd) for interface-specific filters is enhanced. Now, when the TLV decode has errors, the process discards the incorrectly decoded IPC message. PR1164055

  • On PTX Series platforms, when a high-priority clock source (bits-a) goes down, the clock status transits from "locked to bits-a" to "holdover" to "acquiring" to "locked to bits-b". When the bits-a comes up, the clock status reverts from "locked to bits-b" to "holdover" to "acquiring" to "locked to bits-a". PR1168000

  • For PTX Series routers, the IPv6 unilist next-hop member will become "replaced" status on Packet Forwarding Engine (PFE) after interface flapping with IPv6 ND (Neighbor Discovery) timeout. While the problem is happening, routing-table will display all right next-hop status but cannot forward traffic since forwarding next-hop in PFE is in "replaced" status and no longer active. PR1177023

  • FPC might generate a core file when issuing clear threads and show threads simultaneously. PR1184113

  • By default SNMP will cache SNMP values for 5 seconds. Sometimes the kernel will cache these values for a longer duration. PR1188116

  • On PTX Series routers with FPC type 1 and FPC type 2, if there is a problem with ASIC in the FPC, the FPC might be disconnected from the Routing Engine. PR1207153

  • In some conditions where the fan tray is not properly seated in PTX Series routers, the present PIN from the fan tray might not be detected and the fan tray is declared "Absent" in the output for the show chassis environment command. However, the alarm for this condition is not raised under "show chassis alarms" if the alarm occurs during a system reboot. PR1216335

Forwarding and Sampling

  • The Sampling Route-Record Daemon (SRRD) process does not delete routes when the DELETE is received from RPD. This results in build-up of memory in SRRD daemon and once SRRD reaches the limit, it crashes and restarts itself. This scenario only occurs when one family is not configured on all the FPC clients (for example, FPC with inline J-Flow enabled or PIC with PIC-based sampling enabled in one client). Only IPv4 family is configured in all the clients, and IPv6 and MPLS families are not configured for sampling in any of the clients. PR1180158

Infrastructure

  • When the kernel tries to collect statistics from a faulty FPC, it might trigger a kernel panic because of an invalid response from the faulty FPC. PR1185013

MPLS

  • In the following scenario where 1) The PHOP link goes down and the router becomes MP for a LSP. 2) After some time, NHOP link for the same LSP goes down. The router becomes PLR for the same LSP. So effectively, the router is both MP and PLR for the same LSP. In this scenario, the router sends incorrect PathErr message for the backup MP PSB. It sends "Bad strict route" PathErr instead of "Tunnel local repaired" PathErr.PR1132641

  • Changing the configuration under both [ protocols pcep ] and [ protocols mpls lsp-external-controller ] might trigger rpd to crash because of a race condition. PR1194068

Platform and Infrastructure

  • When you configure one group with a configuration of routing-instances and apply that group under routing-instances, the rpd process crashes after executing you run the activating routing-instances or deactivating routing-instances commands. PR1109924

  • In a very rare scenario, during TAC accounting configuration change, the auditd daemon crashes because of a race condition between auditd and its sigalarm handler. PR1191527

Routing Protocols

  • A PTX Series node with a PR 1169289 fix might not be able to play the role of 6PE ingress node for inet6 traffic, if multipath is enabled for the peer giving the inet6 routes in the "inet6 labeled-unicast" family. This problem occurs because PR 1169289 causes the PTX Series router to create a composite next hop for the inet6.0 route, which is not supported. PR1185362

User Interface and Configuration

  • When persist-groups-inheritance is configured and you issue a rollback, the configuration is not propagated properly after a commit. PR1214743

Resolved Issues: 15.1R4

Class of Service (CoS)

  • In case of member links of an aggregated Ethernet (AE) interface scatter over multiple Packet Forwarding Engines, if the FPC where member links of the AE interface reside gets reset or the interface is disabled, there may be a dip in the output of SNMP walk on an AE-related queue MIB (such as jnxCosQstatTxedPkts). The behavior is intermittent and not seen every time. PR1122343

  • This PR does optimization in AE SNMP handling. If all the links in an AE bundle go down, then any CoS SNMP query for this AE IFD/IFL will return cached values. PR1140440

General Routing

  • FFP is a generic process that will be called during commit process, and FFP calls the PDB initialization as part of its process. On the PDB-unsupported platforms (MX Series, EX9200, M10i, M120, M320 is PDB-supported), when committing configuration, some error messages will be seen. PR1103035

  • When a labeled BGP route resolves over a route with MPLS label (for example, LDP/RSVP routes), after clearing the LDP/RSVP routes, in the short window before the LDP/RSVP routes restore, if the BGP routes resolves over a direct route (for example, a one-hop LSP), the rpd process might crash. PR1063796

  • When a switchover is done from one Routing Engine to the other, in graceful-switchover redundancy mode, there is a brief period early in the transition of the SIB to online state, during which unsoliciited (not corresponding to an attempt by the CPU to access the SIB via PCIe) errors are received at the downstream PCIe port on the CB to the SIB. The fix is to mute the generation of such errors during this brief period of the switchover. PR1068237

  • The MIB counter or "show pfe statistics traffic" shows junk PPS and invalid total traffic output counter. PR1084515

  • On PTX Series platforms with external clock synchronization interface configured, when both BITS external clocks are disconnected at the same time, the 100GbE-LR4 FINISAR interface might flap. This link flap issue is narrowed down to the operation of datapath FIFO within CFP. When both the BITS clocks are disconnected, the reference clock jumps to "free-running" mode. This transition leads to a phase shift in the reference clock. Because of this phase shift, the data rates into and out of the FIFO will temporarily not match, leading to a FIFO over-run or under-run condition. This over-run or under-run condition forces a FIFO reset, and the output signal is distorted. So the far-end interface detects a “local-fault,” then returns a “remote-fault” back to the near-end, hence a link flap. Users need to manually configure the FPC recovered clock port for each clock put into "chassis synchronization source". Only one clock of each FPC can be put into "chassis synchronization source". PR1091228

  • On PTX Series platforms, if there are scaling configurations (for example, 5,000 routes and each of them with 64 ECMP paths configured) on a single interface and an L2 rewrite profile is applied for the interface, the FPC might crash when deactivating and then activating the CoS configuration of the interface. PR1096958

  • Entropy Label Capability is enabled by-default on all Juniper Networks (PTX Series and MX Series) systems. On PTX Series routers transit LSRs that carry LSPs with Entropy Label Capability, packet loss can be observed caused by data errors when one or more labeled route entries are not properly removed from the hash table (That is, following LSP optimization or MBB event) because the “stale” entries are pointing to corrupted route memory. As a result, when the MPLS label that is associated with the stale entry is reused, data errors are seen for packets using the corresponding label. PR1100637

  • Because of a buffer size issue for FPC-SFF-PTX-P1-A (PTX3000) and FPC2-PTX-P1A (PTX5000), the "ISSU RECONNECT TIMEOUT" or "READY Message Without Reconnect" message is seen during unified ISSU . PR1155936

High Availability (HA) and Resiliency

  • On MX Series platforms with Junos OS Release 15.1R1 or later, while a core file is being generated, if you try to access the dump file directory, the system might hang and crash due to the deadlock defect.PR1087082

Interfaces and Chassis

  • During subscriber login or logout, the following error log might occur on the device configured with GRES/NSR: /kernel: if_process_obj_index: Zero length TLV! /kernel: if_pfe: Zero length TLV. (pp0.1073751222) PR1058958

MPLS

  • When an LSP is link-protected and has no-local-reversion configured, if the primary link (link1) is down and LSP is on bypass (link2), then another link (link3) is brought up, before the LSP switch to link3. If link1 is enabled and link3 is disabled, the LSP will remain in bypass LSP forever. This is a timing issue. PR1091774

  • When multipoint LDP (M-LDP) in-band signaling is enabled to carry multicast traffic across an existing IP/MPLS backbone and routing process is enabled to use 64-bit mode, the rpd might crash because it is attempting to access an uninitialized local variables. PR1118459

Network Management and Monitoring

  • While the router is rebooting and SNMP polling is not stopped, SNMP requests might land on the mib2d process before Routing Engine protocol mastership is resolved, causing the mib2d process crash. PR1114001

Platform and Infrastructure

  • With the delta-export command enabled, "show|compare" output still appears after the last successful commit. PR1129577

Routing Protocols

  • In an IS-IS environment MPLS LSPs are established, when the IS-IS traceoptions flag "general" is activated, and the LSP convergence time is increased. PR1090752

  • In a multicast environment, when the rendezvous point (RP) is a first-hop router (FHR) with MSDP peers, when the rpf interface on the RP is changed to an MSDP-facing interface, traffic loss is seen. The loss occurs because the multicast traffic is still on the old rpf interface, so a multicast discard route is installed.PR1130238

Software Installation and Upgrade

  • In certain conditions, when /var is not mounted from a persistent file system, executing a Junos OS upgrade will have unexpected results. This is caused by an inexact check of whether Junos OS is running from an Emergency VAR. PR1112334

VPNs

  • For a Layer 2 circuit, the PTX3000 uses a different Virtual Circuit Connectivity Verification (VCCV) BFD control packet format from that of MX Series and the other PTX Series platforms. PTX3000 negotiates the router-alert control channel type and uses the PW Associated Channel Header of Channel Type : 0x0021. However, MX Series and the other PTX Series platforms use the channel Type 0x0007 without IP/UDP headers. Junos OS takes the Channel-type 0x0007 as default. MX Series and the other PTX Series platforms work as expected. This is a PTX3000-specific issue. PR1116356

Resolved Issues: 15.1R3

Class of Service (CoS)

  • In case of member links of an aggregated Ethernet (AE) interface scatter over multiple Packet Forwarding Engines, if the FPC where member links of the AE interface reside get reset or the interface is disabled, there might be a decrease in the output of SNMP walk on the AE-related queue MIB (such as jnxCosQstatTxedPkts). The behavior is intermittent and not seen every time. PR1122343

  • This PR does optimization in AE SNMP handling. If all the links in an AE bundle go down, then any CoS SNMP query for this AE IFD/IFL will return cached values. PR1140440

General Routing

  • When a labeled BGP route resolves over a route with an MPLS label (for example, LDP/RSVP routes), after clearing the LDP/RSVP routes, in the short time before the LDP/RSVP routes restore, if the BGP routes resolve over a direct route (for example, a one-hop LSP), the rpd process might crash. PR1063796

  • When a switchover is done from one Routing Engine to the other, in graceful-switchover redundancy mode, there is a brief period early in the transition of the SIB to online state, during which unsoliciited (not corresponding to an attempt by the CPU to access the SIB through PCIe) errors are received at the downstream PCIe port on the CB to the SIB. The fix is to mute the generation of such errors during this brief period of the switchover. PR1068237

  • On PTX Series platforms with external clock synchronization interface configured, when both BITS external clocks are disconnected at the same time, the 100GbE-LR4 FINISAR interface might flap. This link flap issue is narrowed down to the operation of data-path FIFO within CFP. When both the BITS clocks are disconnected, the reference clock jumps to "free-running" mode. This transition leads to a phase shift in the reference clock. Due to this phase shift, the data rates into and out of the FIFO will temporarily not match, leading to a FIFO over-run or under-run condition. This over-run or under-run condition forces a FIFO reset, and the output signal is distorted. So the far-end interface detects 'local-fault', then return 'remote-fault' back to the near-end, hence a link flap. User needs to manually configure FPC recovered clock port for each clock put into "chassis synchronization source". Only one clock of each FPC can be put into "chassis synchronization source". PR1091228

  • On PTX Series platform, if there are scaling configurations (for example, 5000 routes and each of them with 64 ECMP paths configured) on a single interface and L2 rewrite profile is applied for the interface, the FPC may crash when deactivating and then activating the CoS configuration of the interface. PR1096958

  • Starting with Junos Release 14.1, Entropy Label Capability is enabled by-default on all Juniper [PTX ] systems. On PTX transit LSRs that carry LSPs with Entropy Label Capability, packet loss can be observed due to data errors when one or more labeled route entries are not properly removed from the hash table (i.e., following LSP optimization or MBB event) because the 'stale' entries are pointing to corrupted route memory. As a result, when the MPLS label that's associated with the 'stale' entry is re-used, data errors are seen for packets using the corresponding label. PR1100637

  • FFP is a generic process that shall be called during commit process, and FFP calls the PDB initialization as part of its process. On the PDB-unsupported platforms (MX Series, EX9200, M10i, M120, M320 is PDB-supported), when committing configuration, some error messages will be seen. PR1103035

High Availability (HA) and Resiliency

  • On PTX Series platform with Junos OS Release 15.1R1 and later, while a core dump is in progress, if we try to access the dump directory, due to the deadlock defect, the system might hang and crash. As a workaround, we should not access the "/var/crash" directory till the core dump is complete. PR1087082

Interfaces and Chassis

  • During subscriber login/logout, the below error log might occur on the device configured with GRES/NSR. /kernel: if_process_obj_index: Zero length TLV! /kernel: if_pfe: Zero length TLV (pp0.1073751222). PR1058958

MPLS

  • When an LSP is link-protected and has no-local-reversion configured, if the primary link (link1) is down and LSP on bypass (link2), then another link (link3) is brought up, before the LSP switch to link3, if link1 is enabled and link3 is disabled, the LSP will stuck in bypass LSP forever. This is a timing issue. PR1091774

  • When multipoint LDP (M-LDP) in-band signaling is enabled to carry multicast traffic across an existing IP/MPLS backbone and routing process is enabled to use 64-bit mode, the rpd might crash due to accessing uninitialized local variables. PR1118459

Network Management and Monitoring

  • While the router is rebooting and SNMP polling is not stopped, SNMP requests might land on mib2d process before Routing Engine protocol mastership is resolved, causing the mib2d process crash. PR1114001

Platform and Infrastructure

  • The MIB counter or "show pfe statistics traffic" shows junk PPS and invalid total traffic output counter. PR1084515

  • With delta-export command enabled, "show |compare" output still shows after last successful commit. PR1129577

Routing Protocols

  • In IS-IS environment, MPLS LSPs are established, when IS-IS traceoptions flag "general" is activated, the LSP convergence time is increased. PR1090752

  • In multicast environment, when the RP is FHR (first-hop router) and it has MSDP peers, when the rpf interface on RP changed to MSDP facing interface, due to the multicast traffic is still on the old rpf interface, a multicast discard route will be installed and traffic loss will be seen. PR1130238

Software Installation and Upgrade

  • In certain conditions, when /var is not mounted from a persistent file system, executing a Junos OS upgrade will have unexpected results. This is caused by an inexact check of whether it is running from an Emergency VAR. PR1112334

VPNs

  • For Layer 2 circuit, PTX3000 uses different VCCV (Virtual Circuit Connectivity Verification) BFD control packet format from that of MX Series and the other PTX Series platforms. PTX3000 negotiates Router-alert control channel type, and uses PW Associated Channel Header of Channel Type : 0x0021. However, MX Series and the other PTX Series platforms use the Channel Type is 0x0007 without IP/UDP headers. Junos OS takes the Channel-type 0x0007 as default. MX Series and the other PTX Series platforms work as expected. This is a PTX3000 specific issue. PR1116356

Resolved Issues: 15.1R2

Forwarding and Sampling

  • In PTX Series Carrier-Grade Service Engine (CSE) jflow solution environment, because the sampling process (sampled) may get into a continuous loop when handling asynchronous event (for example, aggregated tethered services interface flapping, or route update, or IFL/IFD update), the sampled may never come out of that loop which may result in high CPU usage (up to 90 % sometimes). Because, sampled is not able to consume any states (such as route updates, interface updates) generated by kernel and this results in memory exhaustion, finally resulting in the router not making any updates and forcing a router reboot. PR1092684

General Routing

  • On PTX Series platform, when performing scaling (for example, polling 768 IFDs via SNMP with max of 92 PPS and with all 8 FPCs online) SNMP polling on the device, due to the large number of messages between Routing Engine and Packet Forwarding Engine, PFEMAN (Packet Forwarding Engine manager) errors might be seen on the router, which may cause high SNMP response time and CPU spike (for example, increase 8 % when executing the "show" command) as well. PR1078003

  • On PTX3000 routers running Junos OS Release 14.1 and later, the Packet Forwarding Engine does not support L3VPN VRF. For example, when you assign the loopback (lo0) interface to VRF as the management VRF, the following commit error is returned: # commit check [edit routing-instances l3vpn interface] 'et-8/0/0.0' RT Instance: Only loopback interface is supported under vrf routing instances. error: configuration check-out failed Note that in Junos OS Release 14.2, you will see the same commit error, but the commit will be successful. You might also encounter a packet discard issue. PR1078960

  • Tunable SFP+ optics will not be supported on P1-PTX-24-10G-W-SFPP PIC in Junos OS 15.1R1 release. On Tunable Optics in this PIC, with 15.1R1, the wavelength will not be configurable and the tunable parameters will not be correctly displayed in the CLI. PR1081992

  • The FPC on PTX Series router might crash and reboot when the Packet Forwarding Engine is handling a fatal error; when the error happened, "TQCHIP0: Fatal error pqt_min_free_cnt is zero" log message will be seen. PR1084259

  • Due to a change in an existing PR, group names in the configuration must be a string of alphanumericals, dashes, or underscores. There is not workaround other than following the group name instructions. PR1087051

  • In Dual Routing Engine systems when both Routing Engines reboot and after coming up, if the mastership is not established or takes time to establish, mib2d may start and exit four times in quick succession. Hence it will not be running. As a workaround, it can be simply started again once Routing Engine mastership is established. This is a race condition and hence may not be seen always. PR1087428

  • On PTX Series platforms, some non-fatal interrupts (for example, CM cache or AQD interrupts) are logged as fatal interrupts. The following log messages will be shown on CM parity interrupt: fpc0 TQCHIP 0: CM parity Fatal interrupt,Interrupt status:0x10 fpc0 CMSNG: Fatal ASIC error, chip TQ fpc0 TQCHIP 0: CM cache parity Fatal interrupt has occurred 181 time(s) in 180010 msecs TQCHIP 0: CM cache parity Fatal interrupt has occurred 181 time(s) in 180005 msecs PR1089955

  • On Junos OS Release 15.1R1, when the multicast next-hop is changed, the grafting and pruning operations take more time than before. PR1090608

  • When the PTX Series only has bits-a and bits-b as configured clock sources (and there is no interface on FPC configured as clock souce), and it is losing signal from both of bits-a and bits-b simultaneously, clock sync state will go to FREERUN mode immediately, this is unexpected behavior. After the fix of this PR, clock sync state will stay HOLDOVER, then will go to FREERUN mode after the timeout. PR1099516

  • On PTX Series platform, when yanking out FPC or SIB ungracefully (for example, pulling the line card out of the chassis unintentionally when the line card is carrying the traffic), there might be small probability that it can impact any of the FPCs with Grant Scheduler (GS) and Request Table (RT) fatal interrupt occurred. PR1105079

Interfaces and Chassis

  • If we load the 15.1 Junos jinstall/jinstall64 image on PTX Series and if we have CFM configured over AE interfaces, the FPC might crash. PR1085952

  • In the dual Routing Engines scenario with GRES and ae0 interfaces configuration, if GRES is disabled on system, the backup Routing Engine should remove the ae0 bundle; however, it does not go clean and ae0 remains in the backup Routing Engine. After switching Routing Engine mastership to make other Routing Engine as master, the new master Routing Engine (which was backup earlier) continues to use invalid MAC address "00:00:00:00:00:00". PR1089946

  • After removing a child link from AE bundle, in the output of "show interface <AE> detail", the packets count on the remaining child link spikes, then if add back the previous child link, the count recovers to normal. PR1091425

  • On PTX Series platform, if the configurations that have per-unit-scheduler configured on the interface, but without proper class-of-service configuration for the same interface, due to lack of commit check, the device control daemon (dcd) may fail to return "commit error" and pass the configuration. Following is an example: user@re0# set interfaces et-0/0/1 per-unit-scheduler vlan-tagging unit 0 <<<<< The configuration for interface et-0/0/1 user@re0# commit check error: per-unit-scheduler is configured but class-of-service is blank <<<<< This is correct behavior error: configuration check-out failed <<<<< .. user@re0# set class-of-service forwarding-classes queue 7 q7 <<<<< user@re0# commit check configuration check succeeds <<<<< This is wrong behavior because et-0/0/1 does not have class-of-service configuration * If reboot this router after committing, the administrator cannot access without console because the router cannot read this configuration. When deleting the above configuration after rebooting, telnet etc could be used. PR1097829

MPLS

  • In the output of the CLI command "traceroute mpls ldp", the addresses of the interfaces on transit PTX Series routers might be shown as "127.0.0.1". PR1081274

Network Management and Monitoring

  • Due to inappropriate cleanup in async library, disabling multiple interfaces while SNMP is polling interface oids might cause mid2d process to crash. PR1097165

Routing Protocols

  • On PTX Series platform with transit BGP-LU chained composite next-hop configured, when advertising LDP routes via BGP labeled unicast (BGP-LU), if the LDP LSP itself is tunneled over an RSVP LSP, the rpd process might crash. Note: The "set routing-options forwarding-table chained-composite-next-hop transit labeled-bgp" is enabled by default on PTX Series. PR1065107