Revision Notes for Router Software 7.71 rev 10
Date: 11/2/94
Subj: Revision Notes for 7.71 fix 9
=========================================================================
SITE MANAGER COMPATIBILITY
=========================================================================
Router version is managed by Site Manager version
v7.71 1.71 1.71 fix 1, 2, 3, 4, 5
7.71 fix 1, 2, 3, 4, 5, 6, 7, 8, 9 1.71 fix 1, 2, 3, 4, 5
Revision fix/7.71/10 corrects the following problems:
=============================================================================
CR 10583: In a dial backup configuration. When the Tx wire is disconnected
8630: from a Slave router, the master's primary line goes down. When
the slave accepts the call from the master a bus error occurs.
======================== Version 7.71 fix 9 =============================
The image for FRE hardware is: 1641794 Nov 1 10:32 boot.exe
The image for VME & AFN Floppy hardware is: 1686444 Nov 1 10:41 ace.out
The image for AN hardware is: 998150 Nov 1 10:44 boot.exe
The image for IN & AFN Flash hardware is 1539900 Nov 1 10:48 boot.exe
_____________________________________________________________________________
CR 10029: When configuring outbound filters based on source or destination
MAC address, the MAX address must be entered in canonical format.
The problem then arises that when entering a MAC address range in
canonical format, the high address range could end up with a lower
value than the low range and Site Manager will give a message
'Minimum Value Greater Than Maximum Value'. Added a new MIB variable
'wfCctOptsEntry.wfCctOptsMacFilterFormat'. The default is 1,
which is canonical. Setting this to 2 will allow entering MAC
filters in token ring format. (Note: All filters on the interface
must be in token ring format!). Site Manager support for this option
will implemented in a future release.
CR 10563: When running Vines over Frame Relay with Vines Fragmentation
Protocol (FRP) disabled, Vines traffic will not be forwarded.
CR 10746: IPX router responds to Get Nearest Server request when sap
filters are configured on the same interface where both a
server and a client reside (10746).
CR 10857: IPX router doesn't update its SAP advertisement after socket
number changes.
CR 10859: IPX router does not correctly forwarded IPX Netbios frames
(type 14h).
CR 10667: IPX Routes May Fail to Age Properly. The Lost Route implementation
incorrectly forms the packet to be sent out in the case where that
route was learned to be dead (hop count of 16) from a different
interface than the one that the route was originally learned from.
CR 10819: XNS can restart on the slot where one or both sync lines in a
multiline configuration go down/up.
CR 9999: Slot can reset with 'Out of Memory' errors when IPX interface with
many (> 400) hosts entries goes down.
CR 10787: IPX router does not age out rip entries received with a hop
count of 15.
CR 10607: EGP neighbour connection not re-established after circuit goes
down/up.
CR 10763: Transparent bridge / Multiline interface fails to come up on boot
or slot reset.
CR 10745: Enhancement for Dial-on-Demand circuits to allow the configuration
of the inactivity timer to be in seconds instead of minutes. Note
that this change affects existing config files such that what the
network administrator previously configured in minutes is now a
value in seconds.
CR 8616: Dial back up circuit is unable to establish the PPP link the first
time the primary link goes down after being booted.
CR 10412: When Learning bridge traffic is flooded over FDDI onto a Frame
Relay network the router at the remote side does not flood the
bridge traffic out its other interfaces which reside on the same
slot as the Frame Relay interface.
CR 9453: When about 300 external networks become unreachable and corresponding
ASE LSAs are flooded with Max Age throughout OSPF AS, OSPF on some
routers (depending upon configuration) may generate Hogs Bufs errors.
CR 10097: Doing a 'stop 6' from TI on a CN causes panics in dmap.
CR 10668: Unable to establish a Telnet connection with AFN - received the
following message in the log `Connection manager refused connection
from , State: 3.'
CR 10696: Learning bridge over token ring does not learn about station
moves as it should.
CR 10009: After the primary line is disabeled, initialization of the
backup dial circuit can cause a panic. The PRIMARY line was
configured as STANDARD.
CR 10314: In particular configuration DECnet router sends incorrect level
2 routing message designating local area as 1 hop and some non
zero cost away.
CR 9958: Vines router does not respond to StreetTalk requests from clients
on serverless segments.
CR 9870: Dynamic disabling of an OSPF interface can cause a PANIC bus
error on non-backbone routers.
CR 10417: BGP-3 Update messages are sent with incorrrect next hop in
particular configuration.
CR 10543: Slot resets with 'System restart in tmo_wdog.c at line 51' when
XNS router learns many (500) routes after wan link goes down/up.
CR 10685: Running NetBIOS Caching may cause panic in sr_nb_process_are due
to RIF entries being aged while the same RIF entries are being
relearned.
CR 10684: Running NetBIOS Caching may cause a 'PANIC in mib_ent.c at line
710' due to RIF entries being aged while the same RIF entries
are being learned.
CR 10379: The 'show bgp peers' command doesn't show configured values for
Hold and Keep Alive timers.
CR 7221: IP and OSPF remain up when FDDI interface is disconnected.
CR 9175: Router responds to SMDS hartbeat poll on logical link 0 instead
of logical link 1.
======================== version 7.71 fix 8 =============================
The image for FRE hardware is: 1641587 Sep 30 12:38 boot.exe
The image for VME & AFN Floppy hardware is: 1685350 Sep 30 12:45 ace.out
The image for AN hardware is: 999610 Sep 30 12:40 boot.exe
The image for IN & AFN Flash hardware is: 1531042 Sep 30 12:43 boot.exe
=========================================================================
CR 9466: The MIB variable 'wfIpInterfaceForwDatagrams' decreases instead
of increasing on OSPF interfaces.
CR 9885: When IP data is to be forwarded to a network which is not directly
connected to the local router and is available via a non-peer
BGP-3 speaker, the router fails to ARP for the next hop media
address of the non peer BGP-3 speaker.
CR 7188: Poor performance (Lack of resource errors dropping packets on
ethernet interfaces) in large (1000 nets, 800 saps) IPX networks.
CR 9968: In particular configuration where certain packets are received
on an ethernet interface in a specific order VINES packets are
lost (due to being delivered to wrong ISAP) causing excessive
delays in file transfers.
CR 10076: Dial on demand using X21 terminal adaptors, ISDN, V25bis
signalling on 44 pins sync interface does not work. Sync
interface cannot see the modem, reports CTS failure. Incoming
calls never transit out of v25bis signalling to data mode, all
LCP configuration frames are interpreted as illegal v25bis
commands. Outgoing calls never transit from modem signalling
to LCP configuration phase.
CR 10273: If there is a default route configured and BGP-3 export filters
as in:
IGNORE default range from ANY protocol from ANY BGP AS from
specific Peer,
PROPAGATE default route (0.0.0.0+255.255.255.255) from ANY
protocol and any AS for specific Peer,
PROPAGATE class B natural network rangefor everything.
Watchdog expiration slot resets can occur in routine
'ip_ck_export_rules' routine when BGP-3 processes default route.
CR 9887: During the learning of BGP-3 routes from 2 BGP-3 peers (both
advertising the same routes) router may generate Hogs Bufs errors
in the BGP-3 module leading to PANIC's from other processes
trying to get buffers.
CR 10047: Access Node Token ring interface does not receive a Source
Route Bridge SRF, ARE, or SRE with a RIF of size greater than
14 bytes.
CR 9690: The External Route Tag Field within an OSPF AS External Link
State Advertisement is inconsistant when the ASE is generated
from a 7 series router. The use of this field, although not
documented in the OSPF specification to this effect, in the 5
series is to list the IP address of the gateway from which an ASE
Type 2 was originally learned via RIP (or other protocol). The 5
series does this consistently and has proved to be an invaluable
troubleshooting tool when debugging a large OSPF network where
there are multiple routers advertising the same ASE Type 2s. The
7 series on the other hand, will use the field in the same way
initially, but seems to lose track of what data is actually being
plugged into the field if the ASE Type 2 is refreshed for any
reason. An example of when this might happen is if an ASE Type
2 is aged, flushed and then relearned and refreshed again. The
External Route Tag field may look fine in an advertisement prior
to and when it is actually flushed, but when a new instance is
generated, bogus data may appear in this field.
CR 9215: The MIB variable wfOspfIfEntry.wfOspfMtuSize added in version
fix/7.71/1 is supposed to control the MTU size for OSPF packets
sent over this interface. When set to a value of 2, it should
send packets no larger than the IP MTU size for Ethernet. This
restriction is not working and much larger packets are being sent.
CR 9978: Force takedown does not work for dial demand and takedown does
not work for dial backup.
CR 9467: The Threshold feature configured under Global Protocols which
allows messages to be put in the log file when certain statistics
reach predefined values, displays the threshold event in the log
with asn.1 notation (such as 1.3.6.1.4.1.18.4.3.1.1.28.2.1)
instead of the variable that it represents (wfCSMACDErr).
This functionality has been provided by a new configuration option
in Statistical Threshold and Alarms (STA). The administrator can
now configure the threshold Object name portion of the MIB
variable to be a user defined string.
For example, if the ThresholdLabel is not configured the logged
event looks like;
'Object 1.3.6.1.4.1.18.3.4.5.1.36.2.1 with value = 233 units/sec
is > low threshold.'
If the ThresholdLabel is configured the logged event looks like;
'Object wfSyncTxOctets.2.1 with value = 235 units/sec is >
low threshold.'
To configure from TI;
s wfStaThresholdEntry.wfStaThresholdLabel.14.1.3.6.1.4.1.18.3.4.5.1.36.2.1
"wfSyncTxOctets.2.1".
With SiteManager version 2.0 or greater it can be configured via
protocols->global->threshold>->wfswconfig->wfline->wfSyncTable->
wfSyncRxoctes, bring up the window to add the port and circuit
number to see the Threshold label screen.
CR 9839: Setting up Bridge Traffic Filters to drop IP broadcast traffic
while bridging and running IP 'Not Forwarding' mode does not
work. The Traffic Filters are not applied.
CR 9842: With LNM enabled on a Token Ring interface receiving a lot of
9982 traffic continuously 'scheduler exited due to orphaned buffer'
faults can occur.
CR 9708: After configuring an ospf import route filter to ignore all type
1 externals, the type 1 externals do not get filtered.
CR 9960: Multiline circuits do not recover properly when there are more than
2 lines per circuit and one of the lines goes down.
CR 9225: Bridge traffic doesn't properly switch back to the primary line
8743: from a dial backup one. The router may stop learning and taking
incoming traffic from the sync port. Also the router may loop
back broadcast packets in the sync port.
CR 9761: When packets are fragmented by X.25 using the M bit, there is a
forced 200ms delay between fragments (even when the window is
open) providing that no inbound packets are seen. This causes a
sub-optimal use of the transmit window which is especially
noticable over a satellite link.
CR 7711: 'PANIC in map_rem.c at line 60' error can occur when dial backup
circuit is changed from primary to normal.
CR 9739: Spanning Tree does not forward packets immediately after changing a
previously blocked port to forwarding when convergence is completed.
CR 9197: IP Host is not relearned after an adjacent host entry for that host
is dynamically disabled. Also, if this host happens to be a router,
then any networks beyond that router become unreachable.
======================== version 7.71 fix 7 =============================
The image for FRE hardware is 1641416 Sep 2 15:42 boot.exe
The image for VME hardware is 1683972 Sep 2 15:49 ace.out
The image for AN hardware is 1001065 Sep 2 15:53 boot.exe
The image for IN hardware is 1530802 Sep 2 15:55 boot.exe
=========================================================================
CR 9849: IP proxy arp does not respond to different networks as it did
in 5 series platform.
Note: The fix for this problem was to add a new attribute
wfArpBaseNonlocalDest into ARP_BASE_GROUP mib record. If
wfArpBaseNonlocalSDest is set to DROP, the arp request will
be dropped if the source and destination are from different
networks/subnetworks; If wfArpBaseNonlocalSDest is set to
ACCEPT, the arp request will be processed. The default option
is DROP.
CR 9866: IP router drops TCP packets when IP option padding bytes are
non-zero.
CR 9811: Token Ring interface (16meg) configured using FAST MAC PLUS
microcode stops receiving source route bridge frames.
CR 9482: Router running OSPF as an AS Boundary Router may generate
'PANIC: exception vector 2 - Bus error' in routine
'add_network_route' when it shares advertisement of a common
AS External Route with an adjacent neighbor.
CR None: Decnet router exhibits poor performance (sometimes including
watchdog PANICs upon circuits going down/up) in large topology
of many (> 15) areas.
CR 9731: SNMP encounters a bus error and PANICs after several Quick Gets
(of attributes 83 thru 90 of wfIpInterfaceEntry) sent from Site
Manager.
CR 9428: If three routers (a,b,c) are connected together in the config-
uration (a--b--c) where (a) is a AS boundary router, (b) is a
area border router and (c) is a router connected to the area border
router, a problem occurs when we sever the connection between
routers (b) and (c). If we look at the ip routing table on
router (c), we will still see an entry for the AS external that
is advertised by router (a). This problem will not occur if all
of the routers are in the backbone. Also the problem will not
occur if another router is attached to router (c).
CR 8743: Learning Bridge traffic doesn't properly switch back to the
primary line from a dial backup line after primary comes back up.
CR 9716: The TI 'save log' command has been improved to record the load
map and image stamp in the log before saving.
CR 9646: When two ABRs bordering the same areas and all members of an area
summary range disappear on one ABR, the configured area summary
range is used for forwarding traffic to the other ABR. (this is
correct). But when the local interfaces come back on the first
ABR, the summary entry is still used for forwarding when it's
not supposed to.
CR 9224: Decnet packets destined for node numbers higher than the configured
value for Max number of nodes (RouteMaxAddr) are not forwarded to
other areas. The configured value for RouteMaxAddr should only
affect routing within an area.
CR 9553: Router token interface configured for IP Host Only/Source Route
Bridge Local ARP request (no rif field) for different IP address
than router causes wfBrSrInterfaceDropInvalidRings to increment.
CR 9468: When running Translating Bridge (TR to Ethernet) in the presence
of an older application, the manner in which we cache (backwards)
and then include a RIF when translating a packet from Ethernet
to Token Ring may cause an application failure.
CR 9301: The get wfKernelEntry.*. command uses excessive amount of
CPU time.
Performance improvements have been made to the way that process
ids are searched. Before this revision (fix/7.71/7), an ACE
booted with the default config file would use 400ms to complete
the get wfKernelEntry.*. command. The same setup now uses
only 20ms.
CR 9185: NetBIOS cache incorrectly stores the RIF from a Name Recognized
frame causing all subsequent packets to be dropped due to a
non-optimal RIF path.
======================== version 7.71 fix 6 =============================
CR none: IPX NetBios static routes may now be used to direct a NetBios
propagation packet (type 20) through a network. This done by
setting a NetBios static route filter for a destination server,
and setting wfIpxBaseNovellCertificationConformance to disable (2).
This MIB attribute must be set to 2 in all routers in the network
for this feature to work. This also requires all NetBios
propagation packets to be sent from a client with the destination
network as zero, unless this packet will pass a static route
filter in that router. (Packets are tested against the static
route filters before the destination is checked. This allows
accepting packets that may not have a destination network of
zero, and will allow for redirecting the packet.)
CR 9330: Booting the router with configuration that contains import and
export route filters will FAULT continously with 'Watchdog
expired, service terminating' message. This problem was introduced
at revision fix/7.71.4.
CR 9477: Learning Bridge sometimes PANICs during momentary period of all
packet buffers on a slot being in use.
CR 8438: IP ARP entry for adjacent host is unexpectedly removed and not
relearned.
CR 8044: X25 has 6 point to point connections to 6 differnet destinations.
If Connection ID 1 and 2 are used to set up call from A to
destination B. Then you cannot use connection ID 1 and 2 to
set up a call from A to Destination C. From a LAN analyzer once
A connects to B connection ID 1 and 2 are being Used. IF A then
tries to connect C, A will report that it already has a call on
connection ID 1 and 2, dropping the connection from C. This
contradicts The Wellfleet Manual which states that Different
destinations can use the same connection ID.
CR 9266: The wfIfEntry.OperStatus toggles between 1,2,3 when an ETHERNET
circuit is disabled.
CR 9029: IPX router configured for Tick based routing responds to getnearest
server requests with a remote file server even though there is a
local file server of the same type on the network.
CR 8828: When an IP All Subnets Broadcast(ASB) is received on a multinetted
interface with ASB enabled, the ASB is being sent out on the wire
again. This can cause problems with other stations (eg. PC NFS),
who see the frame re-sent on the wire sourced from a different
MAC address (the Wellfleets) from their own and declare a security
violation.
CR 9223: Ring Purges and dropped packets are seen frequently when high
traffic rates occur on 4 MB Token Ring interfaces.
CR 8444: When an area is configured as a stub area and the import summaries
is set to no (normal config for a stub area) a router may lose
connectivity with the network if its neighbor does not send out
a database description packet with the default summary route in
it. This happens when a router goes down and is rebooted. The
router will regain its connection when the area border router
floods the area with an update packet every thirty minutes.
CR None: The character delimiter that separates portions of the phone
number used for Switched Services can now be configured. This
will help in cases where the '/' cannot be used. Two new MIB
variables called
wfSwservOutPhoneNumEntry.wfSwservOutPhoneNumDelimiter and
wfSwservInPhoneNumEntry.wfSwservInPhoneNumDelimiter have been
added so that users can enter a delimiter of their choice.
Site Manager support for this option will implemented in a
future release.
CR None: The configuration of IPX NetBios Static Routes has been improved
to allow the use of partial NetBios server names for static
routes. This partial name will act as a wildcard. Any NetBios
packet that matches this partial name will be sent to the static
route. The MIB attribute 'wfIpxNetBiosStaticRouteEntry' may now
contain names of less than 16 characters. When a NetBios packet
is received, it is compared to this static route name. The
comparison is now done by the length of the static route name.
If the static route name is 5 characters, the FIRST 5 characters
of the NetBios name in the packet is compared to the static route.
This provides wildcard like name matching on the ending portion
of the NetBios server name.
Note: A NetBios static route of length zero will match ALL packets.
======================== version 7.71 fix 5 =============================
The image for FRE hardware is 1632643 Aug 2 18:26 boot.exe
The image for VME hardware is 1674161 Aug 2 18:33 ace.out
The image for AN hardware is 997976 Aug 2 18:36 boot.exe
The image for IN hardware is 1522484 Aug 2 18:38 boot.exe
=========================================================================
CR 6985: The configuration of IP Static Routes has been improved to
change the maximum number of static routes to the same
destination to 12. The previous maximum number was 4.
CR 9342: IPX traffic filter for Source address drop filter of 00000000
does not filter packets with a source address of 00000000.
CR 7188: IPX packets to be forwarded enter the router and are sometimes
dropped or forwarded to the wrong mac address or wrong protocol.
CR 7674: AppleTalk router tends to PANIC during momentary period of all
packet buffers on a slot being in use.
CR 8118: RIP import route filters incorrectly apply the import apply
mask which can cause incorrect routes to be installed into
the routing pool.
CR 8599: DECnet level 2 router does not listen to Level 2 packets sent to
the Level 2 router multicast address 09002B020000. This will
cause interoperabilty problem with other vendor routers which
send Level 2 Updates to all routers multicast address.
CR 8198: TCP mismanages clients window and creates out of window traffic.
CR 8807: Dynamically disabling and then enabling the Spanning Tree Bridge
can cause multiple PANIC's.
CR 7376: In a Multi-Area environment OSPF can Hog Buffers for long (2 secs)
periods leading to PANICs in other sub-systems due to no
available buffers.
CR 9067: When IP curcuit goes down on local slot traffic is still sent
from remote slots to the local down interface. 'Net Unreachable'
should be sent by the remote slot.
CR 8689: A router running 7.71(+) configured with OSPF as an AS Boundary
Router may generate an ASE LSA for a downed IP Interface. What
will occur is the router will continue to generate/refresh an
advertisement for the down interface causing all other routers
to install the route assuming its reachable. This can become a
real problem if another ASBR with a higher cost to the same route
exists. A routing loop could form. An ASE for a directly
connected network should ALWAYS be max-aged when the interface
to that net goes down.
CR 8688: OSPF looses ASE routes when 'more recent self-originated
ASE_LSA' are received.
CR 8234: IP router installs a natural network from RIP when it has a
directly connected subnet of that same network.
CR 8389: When IP is configured with dialbackup over Wellfleet proprietary
point-to-point circuit and the primary circuit is PPP
'PANIC: exception vector 2 - Bus error' occurs.
CR 8629: OSPF looses memory upon receipt of PtP LSA.
CR 7959: Attempting to telnet from either a local or remote network
station to the subnet broadcast address causes 'PANIC in
tn_ctl_msg.c line 124'.
CR 8832: A 'bus error' panic occurs when a bridge frame that matches
a drop filter is received on an SMDS circuit.
CR 8154: When doing a named boot from within AN diags, the router will
boot up fine but the results of the command get wfHwEntry.26.*
will always state config instead of the file name used in
the named boot command.
Note: This repair will not work until the AN boot PROMs are updated
*AND* the updated router image is booted.
CR 7591: IP stops working on a token ring interface if the MAC address
is changed dynamically.
CR 8450: Vines IPC packets containing a broadcast class of 0 in the
transport control header are not forwarded.
CR 6178: Excessive Hogs bufs warnings by DECnet router when adjacencies
go UP or DOWN in networks containing large (> 400) numbers of
End Nodes. PANICs can also occur in other subsystems when they
are unable to allocate buffers.
======================== version 7.71 fix 4 =============================
The image for FRE hardware is 1623239 Jul 5 17:42 boot.exe
The image for VME hardware is 1664768 Jul 5 17:54 ace.out
The image for AN hardware is 983694 Jul 5 18:00 boot.exe
The image for IN hardware is 1515938 Jul 5 18:05 boot.exe
CR None: In OSPF AS where there are two Area Border routers to the same
non-backbone area, using ranges might lead to the routing loop.
Routing loop occurs when packet to the unknown destination
(but within configured range) was sent.
CR None: AppleTalk router can bus error in routine at_rtm_map when
AppleTalk router re-initializes.
CR 8585: Dynamically disabling the the LNM server causes Tag violation 2.
CR 7282: wfX25VcRemoteX121Addr is always = (nil). It should contain the
true X121 address for the remote station when VCs are setup.
CR 8293: The MIB variable ifEntry.ifOperStatus is not being returned correctly
for an FDDI interface. The interface is always reported as UP.
CR 7896: The default route is incorrectly used when forwarding to an
unknown subnet in a configuration where other subnets of the
same network are known. Those packets should be dropped.
CR None: Import and export route filters for all IP-related routing
protocols (RIP, OSPF, EGP and BGP) work for EXACT match only
although documentation indicates that ranges will be filtered.
CR None: The device driver for the E1 board has been enhanced to provide
support for CRC-4 option. This will enable an E1 link connected
to a remote end which has a 'LineType' of CRC-4 to interoperate.
The option can be configured with a
'set wfE1LineEntry.wfE1LineType 2' TI command. Site Manager
support for this option will implemented in a future release.
CR None: Source Route Bridge filters based on the RIF field are not
working.
CR 8349: IP Router repeatedly panics with bus error when there are more
than 20 OSPF NBMA interfaces configured over Frame Relay PVCs
on a single physical circuit.
CR 7200: The IP router has been enhanced to provide invalid ARP source
messages in the log. The option can be configured with a
'set wfArpBase.4.0 1' TI command. Site Manager support
for this option will implemented in a future release. the new
log message looks like;
# 2: 06/22/94 13:12:43.578 INFO SLOT 2 ARP Code: 5
Invalid ARP Source: 1.1.1.1 MAC: 0000a20687c1 on 2.1.1.1
CR 8397: Bridge configured for Hybrid mode over Frame Relay stops
transmitting frames. Bridge stats increment but frame stats do
not increment. It seems to happen when a pvc bounces. Recovery
is only achieved by a router reset.
CR 8366: Bridge over Frame Relay using Group Mode floods continuously.
Connectivity is possible but as the number of VC's are increased,
thru-put goes down due to unneccessary flooding.
CR 8010: If a Frame relay circuit configured to run Bridge receives a
bridgable frame before Bridge is completely up on that circuit,
it (Frame Relay) will accept the frame and flood/forward the
frame to other circuits. If a filter is configured on the circuit,
the action will not take affect until the line is completely up.
The result may be that entries are added to the forwarding table
that are not desired and traffic being flooded to multiple PVCs
until bridge completely comes up.
CR 8009: If a filter is configured on a bridge interface to drop a frame,
the frame is dropped (discarded) but the source address is still
learned. This is an undesired function. Frames that a dropped
due to filtering should be treated as if they were not present.
If these addresses are not learned, then filters can be configured
to eliminate temporary network loops that cause the bridge table
to forward to the wrong interface.
CR 8008: If a filter is configured on a bridge interface that is on a
Frame Relay circuit, the bridge learning process does not take
place. If a drop filter is configured to drop frames with source
address b, learning will not take place for frames that do not
have a source address of b.
CR 8007: If a bridge filter is set on a sync passthru circuit the action
is not taken and the message 'Filter hit in sync passthru' is
logged (no matter what type of filter is configured).
CR 8262: When Wellfleet DECNET router is the DR on an ethernet segment
that has a cluster aliased group of VAX's attached, and an end
node attempts to connect to the cluster address, or a VAX in the
cluster other then the routing vax of the cluster attempts to
connect to an endnode, the Wellfleet DECNET router sets the intra
bit and it shouldn't. The end node then believes it can get to
the cluster by directly resolving the mac layer address from the
cluster alias DECNet address. This results in failed communication
between the two devices.
CR 8441: A router running Frame Relay configured in Group Mode can
experience 'PANIC: exception vector 2 - Bus error' when multiple
Inverse Arp requests are received at once.
CR 7938: When IPX router sends reply to a get nearest server request, if
2 or more servers have the same tick value, Hop count should then
be checked. IPX router is not doing this which can cause the
router to reply with a server which is not the nearest one. This
problem only appears when RIP method is configured as Ticks - not
Hops.
CR 7758: After running install.bat the OSPF Area ID remains at the default
value of 0.0.0.0 rather than being changed to the value entered
while running install.bat
CR 8164: The system log contains PPP messages indicating a possible
loop-back condition because we continue to use magic number even
after our config request for magic number is rejected. Log
messages look like;
# 256: 03/30/94 11:24:08.897 WARNING SLOT 2 PPP Code: 5
Peer is using same Magic number, possible loopback on cct 1
# 257: 03/30/94 11:27:07.361 WARNING SLOT 2 PPP Code: 0
The previous event on slot 2 repeated 17 time(s). [Code 5]
CR 7806 YMODEM and XMODEM 1K does not work when doing receives of files
to the router. The error 'DUART detected Overrun error.
Total: x' appears in the log for each block that fails.
Please note that this repair allows YMODEM file transfers to
work but 'DUART detected Overrun error. Total: x' errors may
still occur if the slot running TI is busy with other
processing.
CR 7840: After reboot OSPF router with static route configured fails to
generate an AS external advertisement for the destination
network specified in the static route.
CR 7921: When Bootp Relay Agent Timeout secs field value
(wfBootpRelayIntfEntry.wfBootpRelayIntfSeconds.*) is configured
to zero the router will not forward BootpRequest packets with a secs
field of 0. This means that clients using only this value are not
able to boot through the Wellfleet router.
CR 8110: Previously, a request from Site Manager for a directory listing
of a flash volume with a large number of files (approx 19 or more)
returned no result and left TI_RUI in a hung state. A second
directory request would cause TI_RUI to become uninstantiated and
would result in a loss of the ability to issue any commands to the
router from Site Manager. Beginning at this revision (fix/7.71/4),
the directory request returns no result but TI_RUI does not hang.
This most often occurs when the user has loaded a large number of
the 7.80 script files on the flash volume. In 7.80, when these
scripts are officially released, this issue will be completely
resolved through the implementation of the file system MIB.
Until then, it is recommended that if the script files are used,
they should be stored on a separate flash card so that directory
listings of the volumes containing the configuration files and
boot images may operate normally. Directory commands from the
TI were never a problem.
CR 7474: When a Designated Router (DR) change occurs in an OSPF network
using Non Broadcast Multi Access (NBMA), any priority 0 routers
loose connectivity beyond directly connected network. The 0
priority router on NBMA does not participate in the DR/BDR
selection process. When dead timer expires for the DR on the 0
priority router before it expires on the BDR this problem occurs.
CR None: When Static Route (that is also a Host Route) forwarding address
belongs to the same network (subnet) as one of the router's
interfaces 'r_insert failed' PANICs may occur.
CR 7797: Specific Routed Frames get through the Source Route Bridge even
though a filter is defined to drop frames based on its
destination MAC address. An All Routes Explorer frames carrying
the same destination MAC address are correctly dropped.
CR 7076: Source Route Bridge traffic destined for particular networks is
dropped instead of bridged until the Source Route Bridge is
reinitialized.
CR 7500: Vines clients on serverless segments are not always able to
connect to servers, Wellfleet router responds with ICP error
155 (Network unreachable).
======================== version 7.71 fix 3 =============================
The image for FRE hardware is 1617655 Jun 9 16:52 boot.exe
The image for VME hardware is 1664780 Jun 9 17:14 ace.out
The image for AN hardware is 976820 Jun 9 17:40 boot.exe
The image for IN hardware is 1515523 Jun 9 17:32 boot.exe
CR 7962: Bootp reply packet sometimes contains the wrong gateway address
when transmitted out a multinetted Ip interface.
CR 7526 OSPF fails to properly converge in a configuration which
contains two ABrouters attached to the same non-backbone
area which have connections to point-to-point subnetworks
of the same network in the non-backbone area. Problem would
occur under p-to-p link failure due to lack of a stub network
advertisement being installed locally for the ABrouters own
interface on the p-to-p subnetwork.
CR 7896 In a particular configuration OSPF Area Border router originates
into non-Backbone area SUM_NET_LSA for ranges configured in this
area causing a routing loop.
CR 8148 When OSPF neighbor over Point-to-Point link goes down and OSPF
router generates new Router Link State Advertisement, because
Dead Timer expired, wrong link is added into Router LSA.
CR None When OSPF router with multiple OSPF interfaces initializes or
reconverges after a network topology change IP router may PANIC
in routines 'rtm_net_cons' or 'map_rem'.
CR 7798 A routing loop may form within an OSPF AS when handling
duplicate ASE advertisements from different AS boundary routers.
CR 6790 IP router does not respond to a ICMP echo requests sent to
broadcast IP address.
CR none Router does not respond to ARP request destined to Token
Ring Broadcast address 0xC000FFFFFFFF.
CR 7641 OSPF may panic with 'panic in mib_ent.c' or 'panic in ospf_rtmib.c'
6744 due to duplicate instances in the MIB database.
7475
CR 7589 Vines router does not handle Check Neighbors and Get Neighbors
commands.
======================== version 7.71 fix 2 =============================
The image for harpoon hardware is 1610523 May 25 12:08 boot.exe
The image for VME hardware is 1657181 May 25 16:26 ace.out
The image for PIRANA is 974010 May 25 17:36 boot.exe
The image for IN is 1511314 May 25 18:36 boot.exe
======================== version 7.71 fix 2 =============================
CR 7382 Enabling previously disabled Point-To-Point OSPF link
causes "r_insert failed" log message and
PANIC in rtm_net_cons.c at line 293.
CR 7599 If a router is connected to an ETHERNET LAN and is configured
for SNAP encapsulation in the wfIpInterfaceEnetArpEncaps
attribute field, the endstations can send and receive SNAP
encapsulated frames, but the routing protocol frames, i.e.
RIP, are not sent with SNAP encapsulation. This prevents the
endstations from receiving routing information.
CR none Router may fail with tag violations due to runt fddi pkts.
Occasionally packets arrive that contain only "FC:dstMAC:srcMAC"
fields (13 bytes). These 13 bytes along with whatever other
random data that happened to be in the remainder of the buffer
caused "tag violations" slot resets.
CR 7217 OSPF router may fail to established neighbors over
sync lines.
CR 7573 IPX router may panic in mib_ent.c when relearning routing
information which it knew before.
CR 7325 IPX may panic with "PANIC in ipx_sap.c at line 819."
message while Services are rapidly going down/up.
CR 7695 IPX now supports only the configured encapsulations for any given
interface. Previously SNAP encapsulation support for ethernet and
fddi intefaces was automaticaly enabled along with configured ones.
CR 7638 XMODEM/YMODEM transfers to the router may fail. 'Timeout while
reading sector x' errors occur when using protocols that send 1024
byte packets to the router.
The following are documentation errors in Appendix B
"Out-of-Band File Transfers" of document "Using the Wellfleet
Technician Interface":
1. Any example using "xmodem sbyw" should have a warning
indicating that the "w" option could cause an out of sync
condition with the receiving device. If the transfer fails
to start, try again without the "w" option.
2. As a special case, if you are using the Microsoft
Windows Terminal program to transfer files to the router,
then the "e" option should be added to the router's command
line. The "e" option disables EOT verification.
Example: xmodem rbcwe a:config
3. Any example using xmodem rby should indicate that the
destination volume and filename list should NOT be supplied
on the router. This will cause the router to select the wrong
checksum code and the transfer will fail.
CR 7641 OSPF Router which has originated external or summary LSA has
7756 wrong sequence number in it's MIB structure.
CR 6793 OSPF AS Boundary Router advertises default route as external
with Forwarding Address. Natural network is advertised by
border router as summary network LSA. Wellfleet router didn't
use Forwarding Address in routing calculation, if there is
more specific network advertised in external LSA.
CR 7204 Running communications manager under OS/2 and setting
CM link window size to value other than that configured
for LLC, Wellfleet LLC subsystem fails to negotiate window
size during XID exchange.
CR 7256 PPP may fail to operate correctly between 5.xx and 7.xx
software due to the "Magic number" negotiationes.
CR 7116 If a router is running OSPF and it is DR-other and PING
request comes from his neighbor who is also DR-other and
source IP address in the packet does not belong to one
of the router's own interfaces (but is in router's Routing
Table) and the router does not have an ARP entry for this
host in its ARP table, the router does not respond - it
does not issue an ARP request for the neighbor's MAC address.
CR none The information stored in the log file when a PANIC occures has
been improved to include the CPU and Link Module type, revision
and serial number. The log message will look like the following:
"CPU id 3 rev 26 ser 8127 Link Module id 8 rev 7 ser 1175"
The decode for the CPU id is as follows:
ACE12 2 ACE25 3 ACE32 4
AFN 5 IN 6 AN 16
FRE 256 FRE2 768
The decode for the Link Module id is as follows:
ENET1 1 ENET2 8
SYNC1 16 SYNC1A 17
T11 24 DSE1 32
DSE1A 33 DST416 40
SST416A 41 DST4 42
SST4A 43 SST416 44
STOK416 45 SST4 46
STOK4 47 T12 56
T12A 57 ST1 58
T156K 60 E1 61
ST156K 62 SE1 63
T12n 64 ST1n 65
T156Kn 66 ST156Kn 67
E1n 68 SE1n 69
SYNC 80 SYNC2A 81
CMCFDDI 88 IPHFDDI 89
DT 104 DSDE1 112
DSDE1A 113 ENET 114
DSE2 116 DSE2A 117
SSE 118 SSEA 119
DSDE10BT 120 ENET3 132
DSDE2 156 OLDQENF 160
DENF 161 QENF 162
QEF 164 DEF 165
MCT1 168 DTOK 176
MCE1 184 WFFDDI2M 192
WFFDDI1M 193 WFFDDI2S 194
WFFDDI1S 195 WFFDDI2MF 196
WFFDDI1MF 197 WFFDDI2SF 198
WFFDDI1SF 199 FMDSET 200
FMDST 201 FMDSE 202
FMSST 203 FMSSE 204
FMSSE 204 FNSDSE 208
FNSDSDT 216 FNSDST 217
DHSSI 224 SHSSI 225
ESAFNF 232 ESAFDSENF 233
ESAFSSENF 234 ESAFDENF 235
ESAF 236 ESAFDSE 237
ESAFSSE 238 ESAFDE 239
ANSEDS 1024 ANSEDST 1025
ANSEDSH 1026 ANSEDSI 1027
ANSEDSTI 1028 ANSEDSHI 1029
ANSETS 1030 ANSETST 1031
ANSETSH 1032 ANDEDS 1033
ANDEDST 1034 ANDEDSH 1035
ANDS 1036 ANDST 1037
ANDSTI 1038 ANTST 1039
ANSDSEDS 1040 ANSDSEDST 1041
ANSDSEDSH 1042
CR 7614 External network learned by OSPF was advertised to RIP with
metric of 1, as it is an OSPF network. In fact, this metric
should be equal to learned plus one, because OSPF AS is considered
as one hop.
CR 7615 Reserved sequence number of 0x80000000 was assigned to some
OSPF External LSA. This number can not appear in LSA. Router must
starts his numbering from 0x80000001.
CR 7607 MIB ifIndex inconsistency breaks third party network management
applications.
CR 6833 IP Router does not ignore Host Routes that are illegally advertized
by other routers. (RFC1058 section 3.2 & 3.4.2).
CR 7229 Vines router will cause a client to forward ICP errors.
On a serverless Vines segment, clients will forward vines ICP
error number 155 "Network/Host Unreachable" when the router
broadcasts a response to an RTP Request from a client.
CR 7293 Users should be able to disable the FDDI duplicate address
test (DAT) by setting wfFddiSmtExtEntry.18."inst" to 2.
The router continues to send out NIF Announcement frames every 2
seconds instead of stopping.
========================= version 7.71 fix 1 =============================
The image for harpoon hardware is 1609413 May 5 11:58 boot.exe
The image for VME hardware is 1646844 May 5 12:00 boot.exe
The image for PIRANA is 975397 May 5 12:02 boot.exe
The image for IN is 1504719 May 5 12:06 boot.exe
========================= version 7.71 fix 1 =============================
CR 7476 OSPF may fail to list non OSPF IP interfaces in LSDB table
as ASEs.
CR none DLSw would let XID frames to pass in connection pending state.
CR 7335 DLSw could lose frames received out-of-sequence during
data transfers.
CR none LLC will no longer log ENTER/EXIT busy commands.
CR 6834 OSPF may lose memory while operating over Point to Point links.
CR 7331 Ability to enable/disable FRP header on WAN interfaces for VINES
has been added.
CR 7335 LLC may lose some frames during data transfer.
CR 4419 IP routing may fail when router has network and subnet routes of
7157 the same network in its routing table.
CR 6879 SNAP encapsulated ARP packets are bridged to other interfaces
when it should be processed by IP ARP.
CR 6606 The maximum MTU parameter for Sync Lines is changed from
4500 to 4608. Changed the MAXIMUM and DEFAULT to 4608 for HSSI.
CR 1316 New configuration parameter for OSPF interface is added, it will
allow to change the MTU for OSPF packets on this interface.
The parameter wfOspfIfEntry.wfOspfMtuSize is set to 1 by default
which is setting the MTU size to 4500, value of 2 makes the MTU
of the ethernet size (~1500). The change to take effect the
config needs to be saved and the router rebooted.
CR 7274 ICMP redirect messages for host and for net may be sent with
incorrect code value.
CR 6996 Router may reset reporting corrupted stack in the log.
7347
CR 7283 Frame Relay in wellfleet router does not correctly interoperate
with IBM 3745 FEP (NCP v6 rel 2+) using ANSI Annex D.
CR 5610 Added a configurable parameter 'wfIpBaseDefaultOverSubnetEnable'
to allow a default route for subnets. The default setting is
disable.
CR 4751 AppleTalk router may reset while processing a packet with
datalength of zero.
CR 7316 Dial on demand circuits attempt to call out at the same time,
may fail.
CR 7406 Running Translation Bridge, router may reset while processing
RIF info for 2 packets from the same MAC address received at the
"same" time.
CR 7377 IPX router doesn't recover correctly after switching to dial
backup and then back to primary.
CR 7316 A collision may occur establishing dial on demand circuit.
CR 7272 Dial backup does not allways work spaning multiple slots.
CR 7196 Dial demand circuit not established if port is changed
while circuit is active.
CR 6829 Support or raise dtr V.35 autodial is added.
CR 7442 ESA2X0 (Dual Ethernet) link modules produce parity errors when
used with FRE2 boards.
CR 7346 SNMP packets with community = NULL will cause slot to reset.
CR 7462 SNMP gets to mib-II objects of the MIB tree may result in a
bus error end a slot reset.
======================= release of v7.71 ================================
The image for harpoon hardware is 1612668 Apr 21 13:50 boot.exe
The image for VME hardware is 1647823 Apr 21 13:48 boot.exe
The image for PIRANA is 972213 Apr 21 13:53 boot.exe
The image for IN is 1510549 Apr 21 13:55 boot.exe
======================= release of v7.71 ================================
========= this problems have been resolved since 7.70 release ===========
CR 6525 Router with a E1 interface may fail to indicate that service
became unavailable when the interface is disconected.
CR 7068 Running redundant links between two routers configured as
OSPF point to point do not always choose the correct path.
CR 7153 Using DLSw to pass heavy traffic between two routers will cause the
system to go into flow control. Poor performance will be observed,
characterized by the data flowing in spurts, with pauses between.
CR 7092 Net2 compliance test for X.25 interfaces may fail when frame with
a bad CRC is not dropped.
CR 6987 VME routers upgraded to latest 7.xx software may fail named boot.
CR 7002 DLS would not properly queue frames which ware received after
entering the busy condition.
CR 7066 FDDI Link Interface may not attach to the ring under certain
timing conditions.
CR 6919 MIB inconsistances between various Wellfleet products were
6932 corrected.
CR 6791 Invalid interrupts may get enabled on a FDDI interface.
CR 5728 Arp packets originated on one network and destined for different
network shold not be processed by a router.
CR 7173 TCP generates ACK pakets that are not updated.
CR 7160 Bridge input stats count are not accurate and dropped count is
misleading.
CR none Format 7 exception frames will be interpreted correctly.
CR 7272 Dial Backup does not work with multiple lines in pool.
CR 7196 Dial demand circuit not established after changing ports.
CR 7257 OSPF Export Route Filter rule which forced an advertisement to be
a certain type would not work after OSPF advertised the route for
the first time. It would happen after an interface is taken down/up.
CR 6564 VINES routers performance drops 50% sometimes.
CR 7247 AppleTalk router may reset with a tag violation while processing
a malformed NBP packet. If the length of the zone field indicated
larger than 32 it may cause the problem.
CR 7252 TI: cutenv command is not releasing memory used for argument
list.
CR 5396 Decnet router may reset after a router hello packet is received
from a neigbour router which had changed its router priotrity
dinamicly.
CR 7154 FRE2: Parity errors may be reported as TAG violations.
CR 7061 FRE2: Diags are invoked when reducing Global Memory from
16meg down.
CR 6660 IPX NetBios packets with a bad length ware dropped by the
routed without recording. wfIpxHdrErrors will be incremented
if the frame is too short to be a NetBios IPX packet.
CR 4682 A log message has been added to reflect the fact that a
clearlog command was executed from TI interface. This should
be the first entry in the log after it was cleaned.
CR 6265 ARP request packets are sent out as All Route Explorers (ARE) by
Source Routing Bridge. Some RS6000 do not accept Broadcast AREs.
CR 5141 Watchdog timeouts may occur in the IP route cache gate while
processing the forwarding table. Watchdogs can occur while aging
entries in the table, while resetting a range in the table, and
while flushing the default route.
CR 6655 OSPF subsystem may reset when an interface to an area on a
area border router goes down.
CR 7360 SNMP is not getting loaded when 7.71 is booted using 7.56
or 7.57 config files. This would prevent the Site Manager
from accessing the router.
=========================================================================
Feedback | Service Search
| ERC
Copyright
© Bay Networks,Inc., 1997. All rights reserved.
|
/tr> |
