STREAMS T_CONN_IND messages in Solaris 10

We have a STREAMS module that we push on top of TCP to intercept both inbound and outbound connections. We have recently discovered that our module no longer seems to work on Solaris 10. It appears that for inbound connections (i.e. those coming in from the wire), the T_CONN_IND messages do not seem to be passed to our module's rput routine. Anyone know what gives? We know that TCP and IP have been merged in Solaris 10, but we didn't think that would affect modules pushed on top of TCP. Is one of the optimizations that was made to skip the STREAMS messaging for connection indications and do something more direct between TCP and sockmod?
Any help would be appreciated.
Thanks,
Tony

janp2 wrote:
I hit this issue right now so I'm pasting a part of my reply to another list.
It might help other people:
==
The "Warning: Your password has expired, please change it now." comes in a
separate SSH_MSG_USERAUTH_INFO_REQUEST packet. This packet in general has an
"instructions" field, and some "prompt" fields. SunSSH server sends the warning
in a separate info-request packet, with 0 prompts, and with the warning message
in the instruction field. That's fine according to the spec (rfc 4256):
The num-prompts field may be `0', in which case there will be no
prompt/echo fields in the message, but the client SHOULD still
display the name and instruction fields (as described below).
however, when the number of prompts is 0, putty ignores the instruction field.
SunSSH client does the right thing, OpenSSH client as well. In theory, we could
put the warning message in the next info-request packet together with the "New
Password" prompt but that decision was intentional, we would be really "fixing"
stuff to workaround problems somewhere else.
so, my conclusion is that they should file a bug againt those SSH clients they
use. It's not a problem in the SunSSH server at all.
==
BTW, the putty's file is ssh.c, the instruction field is set on line 7474:
s->cur_prompt->instruction = ...
but add_prompt() function that prints the stuff out is called inside of the
following loop:
for (i = 0; i < s->num_prompts; i++)
so, as we can see, if the number of prompts is 0, we get no instruction
field printed.
Jan.Thank you very much, Jan. That was very helpful - at least now I have confirmation.
Update: I contacted the dev team for PuTTY and they let me know that this bug has been fixed in the latest Development version of PuTTY but there isn't a firm release date planned. Anyway, it'll be fixed in the next release of PuTTY
Edited by: dubitancy on Jan 7, 2010 6:33 AM

Similar Messages

  • Example of Streams for Messaging?

    Does anyone have a list of setup steps (maybe scripts?) they would be willling to share that use streams to trap LCR DML changes on a table and place the messages in a queue that will be read by a Java (possibly JMS) process?
    All the examples I can find (which are few including the terribly lacking Oracle docs) are for replicating data to another database. I don't need to do that. I only need to put LCR records in a queue to be read by a Java application.
    Environment is 10g (10.2.0.1) on Solaris.
    Any assistance would be appreciated.

    That's the one I've been trying to work with but the setup also creates replication between two databases. I'm trying to find an easier way to figure out which of those pieces I don't need.
    Some things have changed in 10g as well, an this is a 9i example.

  • Java System.out Message on Solaris

    Hi !
    Could somebody tell me, how I can forward all System.out Messages from
    my Java modules (EJB, Servlets) to the Solaris Terminal window?
    Thanks Eddie

    You could do a tail -f on the kjs console log. (Assuming that you only have one KJS).

  • CPU message on Solaris 8 install on Sparc U30

    When installing Solaris 8 on Ultra Sparc 30, I get the following message after "Configuring /dev and /devices"
    panic[cpu0]/ thread=2a1001dd40: BAD TRAP: type31 rp 2a1001cf30 addr=2f150fcb998 mmu_fsr=0
    followed by;
    sched: trap type=0x31
    followed by what appears to be a memory dump
    followed by Watchdog Reset and Externally Initiated Reset
    Is this a media problem or hardware falilure?

    What's version of NES/Iplanet you are running?
    thanks
    Jong
    "Ed Itkin" <[email protected]> wrote:
    >
    I'm getting strange message when Netscape Client starts on Solaris 8:
    cbfe-dev-web01:/opt/weblogic:263> /opt/netscape/netscape
    ERROR: ld.so.1: /opt/netscape/netscape: fatal: relocation error: file
    /opt/netscape/plugins/libproxy.so:
    symbol __nsapi30_table: referenced symbol not found
    Cant load plugin /opt/netscape/plugins/libproxy.so. Ignored.
    Can you give me idea what happened to NSAPI ? It worked before.
    I rebooted the Solaris server - it did not help.
    Just a reminder : libproxy.so was copied from
    Weblogic 6.0 SP2 ../lib/solaris directory into /opt/netscape/plugins.

  • Messaging in solaris 10

    i have a customer setup with 490 server with solaris 10 ( all patch included) and suncluster3.2. customer wants all console messaging ( dev/sysmsg) to be included into system message(/var/adm/messages) or to be redirected to any other file. i have tried with /etc/syslog.conf or consadm. does anyone who can suggest me on the same ?
    thanks in addvance.
    skd

    Hi Shane,
    Thanks for reply.
    The output of "./imsimta version:
    +Sun Java(tm) System Messaging Server 7.0-0.04 32bit (built Jun 20 2008)+
    +libimta.so 7.0-0.04 32bit (built 01:01:00, Jun 20 2008)+
    +Using /opt/sun/comms/messaging/config/imta.cnf+
    +SunOS mail 5.10 Generic_120012-14 i86pc i386 i86pc+
    “Internal mail” means all domains which have created in our server and when we send mail within domain or in same domain, UID must be validated.
    Thanks
    Prem

  • Error message on Solaris about JNIEnv *env

    Hi,
    I have this line of code in my simple C++ implementation of the native method:
    JNIEXPORT void JNICALL Java_jipclinker_jipc_1ne_1command
    (JNIEnv *env, jobject obj , jstring command)
    const char str = (env)->GetStringUTFChars(env, command, 0);
    // the rest of the code
    when I compile it in Solaris 2.8 I get the following error message:
    In function `void Java_jipclinker_jipc_1ne_1command(struct JNIEnv_ *, class jobject *, class jstring *)':
    16: base operand of `->' is not a pointer
    Can I have peoples ideas about this please?
    Thanks,
    S.

    Try
    env->GetStringTUFChars....

  • NSAPI (libproxy.so) Error message on Solaris 8

    I'm getting strange message when Netscape Client starts on Solaris 8:
    cbfe-dev-web01:/opt/weblogic:263> /opt/netscape/netscape
    ERROR: ld.so.1: /opt/netscape/netscape: fatal: relocation error: file /opt/netscape/plugins/libproxy.so:
    symbol __nsapi30_table: referenced symbol not found
    Cant load plugin /opt/netscape/plugins/libproxy.so. Ignored.
    Can you give me idea what happened to NSAPI ? It worked before.
    I rebooted the Solaris server - it did not help.
    Just a reminder : libproxy.so was copied from
    Weblogic 6.0 SP2 ../lib/solaris directory into /opt/netscape/plugins.

    What's version of NES/Iplanet you are running?
    thanks
    Jong
    "Ed Itkin" <[email protected]> wrote:
    >
    I'm getting strange message when Netscape Client starts on Solaris 8:
    cbfe-dev-web01:/opt/weblogic:263> /opt/netscape/netscape
    ERROR: ld.so.1: /opt/netscape/netscape: fatal: relocation error: file
    /opt/netscape/plugins/libproxy.so:
    symbol __nsapi30_table: referenced symbol not found
    Cant load plugin /opt/netscape/plugins/libproxy.so. Ignored.
    Can you give me idea what happened to NSAPI ? It worked before.
    I rebooted the Solaris server - it did not help.
    Just a reminder : libproxy.so was copied from
    Weblogic 6.0 SP2 ../lib/solaris directory into /opt/netscape/plugins.

  • Stream error message

    Hi Froends,
    I review alert file at bduml and find the Ethereal Network ANalyzer error file.
    The message is as "The file "D:\oracle\product\10.2.0\admin\hqdata\bdump\hqdata_moo1-5700.trc" isn't a capture file in a format Ethereal understands.
    WHat is means for this message?
    How to fix it?
    Thanks,
    JIm

    Jim,
    Is that really related to Oracle Streams? As far as I understand Ethereal is a third party tool and it does not have any relation to Oracle Streams. Could you let us know the contents of the file: hqdata_moo1-5700.trc?
    Thanks,
    Rijesh

  • Error message on Solaris booting

    It was troubled, I have a Question.
    When starting Solaris2.5.1(x86), the following error message is always displayed.
    In order not to display this message, what should it carry out?
    Moreover, what is a cause to display?
    Cause is Hardware? or software?
    -------boot message-----
    Hostname : TEST01
    The file system (/dev/rdsk/c1d0s0) is being checked.
    /dev/rdsk/c1d0s0: INCORRECT BLOCK COUNT I=1053783 (2 should be 0) (CORRECTED)
    /dev/rdsk/c1d0s0: FREE BLK COUNT(S) WRONG IN SUPERBLK (SALVAGED)
    /dev/rdsk/c1d0s0: 25250 files, 419799 used, 1833112 free
    /dev/rdsk/c1d0s0: (3424 frags, 228711 blocks, 0.1% fragmentation)
    The system is coming up. please wait.
    -----end----------------
    Usually, "init 0" (or "shutdown -y -g0 -i0") is used when shutting off a power supply.
    In this case, an error message(incorrect block ,,,,,) is displayed.
    But often, when I use "halt" , an error message may not be displayed
    It cannot repair even if it uses the "fsck" command.
    (Ex. fsck -F ufs -o b=32 /dev/rdsk/c1d0s0)
    I have no idea about a cause.
    please let me know If there are informations.
    ----system environment----
    Solaris 2.5.1 (for Intel)
    Pentium MMX 233MHz, 64MB memory, NIC:3C509B,
    SCSI:AHA-1510B, Serial Multiport:aurora 8000P,
    HDD: WD200BB (20GB, IDE)

    Most likely, this is nothing to worry about. The message is fairly normal. There were some problems with sync'ing disks when such old Solaris x86 systems shut down, but it gets cleaned up when the system boots (which is what the message is telling you).
    If reliability is important to you, upgrade to a modern version of Solaris and turn on file system logging.
    Richard

  • NFS mount error messages on Solaris 8; is there a patch?

    We recently purchased an EMC Celerra NS80 to serve as a front end to our Centerra archive solution. On a Solaris 8 box I've been seeing a large number of NFS errors in the /var/adm/messages file relating directly to the datamover on the NS80. I've tried everything I can think of from the array side of the house to no avail. EMC states there is fix for this problem, patch 113318-12, but according to the patch notes this is for an NFS problem with Solaris 9.
    The error is causing sporadic performance issues when our end users attempt to pull up the data that resides on the Celerra NFS mount, so I would really like to get it resolved. The specific error message is:
    NFS server DATAMOVER1 not responding still trying
    NFS server DATAMOVER1 ok
    The messages of not responding and ok tend to relate every few seconds to 3-5 minutes delay. Our database vendor insists something on the NS80 is introducing this error, but I cannot find anything to change to clear this up.
    Any input would be greatly appreciated.
    Thanks.

    SunOS www02.unix 5.10 Generic_127128-11 i86pc i386 i86pc
    10:58am up 22:51, 4 users, load average: 2.16, 2.26, 2.26
    /export/www
    cache hit rate: 99% (111069900 hits, 6674 misses)
    consistency checks: 459000 (458945 pass, 55 fail)
    modifies: 0
    garbage collection: 0
    /export/zero
    cache hit rate: 94% (2089349 hits, 110629 misses)
    consistency checks: 1497690 (1497075 pass, 615 fail)
    modifies: 0
    garbage collection: 0
    /export/saba
    cache hit rate: 97% (7677577 hits, 174056 misses)
    consistency checks: 10809059 (10801491 pass, 7568 fail)
    modifies: 0
    garbage collection: 0
    So 1 day uptime, that is much better. We rebooted it for mirroring setup, and found that cachefs needs to fsck before it comes up. Can I not simply have it start afresh rather than attempt to keep cache directory? (Obviously I can, but I mean in a boot-friendly manner)

  • Error message while Solaris 10 installation

    I got that error message while installation without knowing the actual reason behind it :
    Searching for configuration file //cdrom/solaris_10/misc/install_config/sysidcfg
    Could not find file.
    Searching for Jumpstart directory : not found
    could not find matching rule in rules.ok

    OK, reinstalled XP, and problem solved.

  • How to create a popup with a messages in solaris 8?

    I want to create a simple popup with a OK button when my users logon to the CDE environment informing them about some important stuff.
    How can I do this the easiest way?
    /C

    If your messages are short enough, you can use the existing /usr/dt/bin/dterror.ds script to display the message you want:
    /usr/dt/bin/dterror.ds "Message" "Title Bar" "OK Button Text"
    If you need more than that, you could adapt that script to your needs. Here's the dtksh reference:
    http://docs.sun.com/app/docs/doc/806-2912

  • Luactivate error message re: Solaris Volume Manager

    Is this something I need to be concerned about?
    2009-04-02 16:40:18 # luactivate sol10_1008_stage2
    ERROR: Solaris Volume Manager, (Root) file missing: </dev/md>.
    ERROR: Solaris Volume Manager, (Root) file missing: </dev/md>.
    Activating the current boot environment <sol10_1008_stage2> for next reboot.
    The current boot environment <sol10_1008_stage2> has been activated for the next reboot.
    2009-04-02 16:40:25 #Thanks,
    Mark

    Is this something I need to be concerned about?
    2009-04-02 16:40:18 # luactivate sol10_1008_stage2
    ERROR: Solaris Volume Manager, (Root) file missing: </dev/md>.
    ERROR: Solaris Volume Manager, (Root) file missing: </dev/md>.
    Activating the current boot environment <sol10_1008_stage2> for next reboot.
    The current boot environment <sol10_1008_stage2> has been activated for the next reboot.
    2009-04-02 16:40:25 #Thanks,
    Mark

  • Inetconv gives error message on Solaris 10

    Hi All,
    On executing the command inetconv -i /etc/inet/inetd.conf i get the following error message -
    incorrect or inconsistent fields : service-name protocol.
    Even after the message it generates the SMF file correctly . I am unable to understand the cause of that error message. Also the data in the inetd.conf is proper.
    Thanks.
    Regards,
    Hemil

    That error is produced when valid_props() in lib/libinetsvc/common/inetsvc.c returns false. It looks like you'll have to examine that code to determine the exact problem. Or you can post the offending line of inetd.conf here, or you can contact Sun Service.

  • Solaris 10 TPI interface

    Hi all,
    I am writing a streams driver to catch the TCP data in between Client/ Server applications under loopback mode using TPI interface. To identify the new listener queue and the client end point information I am using T_CAPABILITY_ACK, T_CONN_RES and T_CONN_IND messages.
    On Server socket (queue) I process the T_CONN_RES messages for ACCEPTOR_id and SEQ_number information, next by finding a matching ACCEPTOR_id field from a lookup table of ACCEPTOR_id, Queue pair (which I have constructed using T_CAPABILITY_ACK message information), I can identify the actual queue which is the listener for further data transfer.
    Also by matching the SEQ_number of T_CONN_RES with the SEQ_number got from earlier T_CONN_IND messages I can identify the IPaddress and port number information of the client end-point.
    When I tested the entire logic on Solaris 10 both on update 2 it works perfectly alright, but the same logic seems to be not working on Solaris 10 initial release.
    In this early release of Solaris 10 all the T_CONN_RES messages are always having Zero ACCEPTOR_id and a constant SEQ_number (always 40).
    The server application and client application both are 64 bit applications running on 64 bit Solaris 10. The Box information is,
    #uname -a
    SunOS devsol14 5.10 Generic sun4u sparc SUNW,Sun-Fire-V240
    #Cat /etc/release
    Solaris 10 3/05 s10_74L2a SPARC
    Is there anything to be done for this release of Solaris, or do I need to catch some additional messages to make sure this works in this release?
    If there is any explanation not clear enough please ask me for more details...
    Any help will be greatly appreciated.
    Thanks in advance!!!
    Regards,
    Laxmi

    If you have installed all required packages, all you'll need to do is wait... Desktop login starts about 5-10 seconds after console login apperas unless you log into console, in which case it won't appear at all.
    I also suggess checking whether SUNWdtlog package is installed since startup scripts for dtlogin are there. Besides if you have the full software group installed you will have a choice whether to use CDE or JDS (Gnome)

Maybe you are looking for

  • 360 wireless received not being detected by Windows 7 64bit

    Greetings, I bought this controller a couple of years ago and it's worked fine with Win7 until now. Now the receiver isn't being detected by 2 different computers. This makes it sound like the receiver has failed, but because it's not really and acti

  • How much do I need to by an iMac 27inch

    How much do I need to by an iMac 27inch ?

  • API to Apply Credit Memo to a Deposit

    We have a requirement to apply the Credit Memo to an outstanding Deposit in Oracle Financials through a program. Can someone provide me 1. An API to Create a Credit Memo 2. An API to apply the Credit Memo to a Deposit Thanks Vasu

  • Won't publish /asks for password/ won't log in .mac

    My website was working great. I changed my password through inspector. Then when I tried to publish again it put me in .mac sign in page. In the past, iweb usually would just publish without going to the sign in page. I wnt ahead and reentered passwo

  • Create Contact Person in SAP SRM 701

    Hello Experts, I'm new in SRM and having a problem with creation of contact person in SRM 701. I use NWBC to access it. Somehow after clicked menu Edit Contact Person (my user id have several administrator role actually and can't find menu related wi