SEARCH CONTACT US SUPPORT SERVICES PRODUCTS STORE
United States    
COMPAQ STORE | PRODUCTS | SERVICES | SUPPORT | CONTACT US | SEARCH
gears
compaq support options
support home
software & drivers
ask Compaq
reference library
support forum
frequently asked questions
support tools
warranty information
service centers
contact support
product resources
parts for your system
give us feedback
associated links
.
} what's new
.
} contract access
.
} browse patch tree
.
} search patches
.
} join mailing list
.
} feedback
.
patches by topic
.
} DOS
.
} OpenVMS
.
} Security
.
} Tru64 Unix
.
} Ultrix 32
.
} Windows
.
} Windows NT
.
connection tools
.
} nameserver lookup
.
} traceroute
.
} ping
OpenVMS VAXPHV10_U2055 VAX V5.5-2 Ethernet Drivers ECO Summary

Copyright (c) Digital Equipment Corporation 1994, 1995. All rights reserved. OP/SYS: OpenVMS VAX COMPONENT: Ethernet Drivers (ECDRIVER.EXE, EFDRIVER.EXE, EPDRIVER.EXE, ESDRIVER.EXE, ETDRIVER.EXE, EXDRIVER.EXE, EZDRIVER.EXE, FCDRIVER.EXE, FXDRIVER.EXE, NET$CSMACD.EXE, NET$FDDI.EXE, XEDRIVER.EXE, XQDRIVER.EXE) SOURCE: Digital Equipment Corporation ECO INFORMATION: ECO Kit Name: VAXPHV10_U2055 ECO Kits Superseded by This ECO Kit: VAXPHV09_U2055 (CSCPAT_0606), VAXPHV08_U2055, VAXPHV07_U2055, VAXPHV06_U2055 ECO Kit Approximate Size: 1000 Blocks Kit Applies To: OpenVMS VAX V5.5-2 Only System Reboot Necessary: Yes ECO KIT SUMMARY: An ECO kit exists for the Ethernet Drivers (ECDRIVER, EFDRIVER, EPDRIVER, ESDRIVER, ETDRIVER, EXDRIVER, EZDRIVER, FCDRIVER, FXDRIVER, NET$CSMACD, NET$FDDI, XEDRIVER, XQDRIVER) on OpenVMS VAX V5.5-2. This kit addresses the following problems: Problems Addressed in the VAXPHV10_U2055 Kit: o This kit and the VAXLAVC04_U2055 kit have the XQDRIVER.EXE image in common. The VAXLAVC04_U2055 has been issued with new images and fixes. The VAXPHV10_U2055 kit is being re-issued with the new XQDRIVER.EXE image to keep it current with the VAXLAVC04_U2055 kit. Problems Addressed in the VAXPHV09_U2055 Kit: o Clarifications were made to the documentation in the VAXPHV08_U2055 Kit. Problems Addressed in the VAXPHV08_U2055 Kit: o NETACP, or any other issuer of IO$_SENSEMAC QIOs, may hang if the DEMFA is experiencing command completion errors. o When a user with Receive All Multicast packets enabled updates the multicast received counter, the system may crash. o A NETDLLERR bugcheck system crash will occur if a P6 parameter is passed to a QIO for the FXDRIVER. o When DECnet Phase V Routing gets an error trying to start a LAN device, the system may crash due to pool corruption. o If a large value is passed for the FMT parameter (i.e., Ethernet, 802), the LAN driver may crash the system trying to access memory that does not exist. o An unprivileged user can consume all of non-paged pool by starting a protocol type in shared mode. When there is not more non-paged pool available, the system will hang. o When a channel is started and an illegal value for the FMT parameter is specified, the high order word of the second longword of the IOSB is set to a random value. This problem is only visible to customers writing their own applications. o Remote Console connection to a DECbridge 90 with NCP may fail when showing port information. A "Target does not respond" message will be displayed. This failure only occurs when NCP is directed to use an FDDI device to connect to the DECbridge 90 on the Ethernet. Problems Addressed in the VAXPHV07_U2055 Kit: o The VAXPHV06_U2055 kit fails to install. Problems Addressed in the VAXPHV06_U2055 Kit: o When multiple users enable the same GSAP, the first user who opens a channel to the device does not receive the packet. o Third-party software to allocate a free SAP on an Ethernet device loops with QIOs for SetMode and Startup until a free SAP is found. The software then exits with quota exceeded and the process BYTLM quota is not credited. o When more than 10 multicast addresses are enabled by the users of XEDRIVER, messages received for addresses after the first ten are not received. o With a MV3100E system running and Ethernet turned on, disconnecting the Ethernet terminator for a period of 7 minutes or longer causes the system to hang. The only way to unlock or release this system hang is to power down the system and reboot. o A system may crash in Ethernet with an INVEXCEPTN bugcheck during promiscuous mode operations if non-paged pool is exhausted. o XQDRIVER (for both the DELQA and the DEQTA) is not reporting the correct number of receive buffers to PEDRIVER. The number of receive buffers for the DEQTA was increased to match the number used for the DELQA. o There is a problem with AST quota depletion. It occurs in the following way: - When a thin-wire Ethernet cable is disconnected and transmits were queued to the device, the device will notify the Ethernet driver of a transmit time-out. The Ethernet driver returns any reads queued to the application with an IOSB status of SS$_TimeOut. Any subsequent QIO Reads are returned with a status of SS$_DEVINACT. - A Customer application then deassigns the channel, assigns another channel and attempts to restart the device. The application ignores the SS$_TimeOut status returned in the IOSB from the restart and posts ten QIO Reads with the specified AST routines. These QIOs are returned with a status of SS$_DEVINACT which causes the deassign/assign loop to be repeated. - When the Ethernet driver aborts the QIO with a status of SS$_DEVINACT, the AST quota for the process should be credited. It is not. The software eventually exits with %SYSTEM-F-EXQUOTA, exceeded quota and the process AST quota is not credited on image exit. o The Ethernet device hardware address returned in a sense-mode QIO is incorrect. o The timer routines in XQDRIVER are not synchronized for SMP systems. Since these timer routines transmit a packet every 8 to 12 minutes, there is a chance that the transmit code will execute simultaneously on multiple CPUs and cause problems that will eventually crash the system (usually with an access violation due to a corrupt transmit pending queue). o Non-privileged user code can shut down the LAVC port. It simply has to $ASSIGN a channel, and $QIO SETMODE! IO$M_SHUTDOWN. o The Ethernet adapter may stop receiving packets, especially if the Lance chip must be stopped and started frequently. Transmits continue to occur normally. o When the FDDI cable is disconnected, FXDRIVER resets the DEMFA every 5 or 300 seconds. After the reset, it reads the SPST register at 1 second intervals to determine if the self test is complete. It should wait 10 seconds before reading any of the device registers or risk an XMI error that would crash the system. o Invalid addresses returned to FXDRIVER cause system crashes. The following errors may occur: - The system crashes with ASYNCWRTER bugcheck with XMI devices. - The Error logger reports XMI Byte count errors. - Boot enabled, Claim token yield and Upstream Neighbor Time-out flags do not have valid values when displayed by SDA. o AppleTalk[R] fails to start after trying 254 unique addresses. The AppleTalk Address Resolution Protocol (AARP) probe message always gets a response if the node number to be used is already in use. The broadcast message it is sending the AARP probe to is enabled at the station. Half Duplex mode in the DEUNA was cleared, causing a copy of the message to be returned to XEDRIVER. NETBIOS also exhibits the same problem. o The following DECnet event is continuously logged on the console. DECnet event 0.7, aborted service request %MOM-E-BADMOPFCT, Bad MOP Function received from target. This problem is caused by the LAN drivers when these three conditions exits on the LAN: 1. A CISCO router is installed, 2. DECnet Service is enabled, and 3. A promiscuous-mode user is enabled on the LAN device. o Several problems were found that track counters, events and time logs. Specifically, these problems are: - The UUB event is not being declared. - The time of the last receive for QIO and ALTSTART users is not being recorded. - The MNE counter is not being zeroed when requested. - The multicast counters in the UCB are not being kept. - The UUB counter for the user and the device are being counted in the wrong field. - The transmit user counters are being incremented for error cases, and they should not be. o The P5 buffer returned from the Ethernet driver is not formatted correctly for the message type received when in promiscuous mode. The test for protocol type should use an unsigned test instead of a signed test. INSTALLATION NOTES: In order for the corrections in this kit to take effect, the system must be rebooted. If the system is a member of a VAXcluster, the entire cluster should be rebooted. [R] AppleTalk is a registered trademark of Apple Computer, Inc.



This patch can be found at any of these sites:

Colorado Site
Georgia Site



Files on this server are as follows:

vaxphv10_u2055.README
vaxphv10_u2055.CHKSUM
vaxphv10_u2055.CVRLET_TXT
vaxphv10_u2055.a-dcx_vaxexe

privacy and legal statement