I wouldn't exclude that possibility at all. Basically there seems to be a problem with packets getting lost somewhere between Sampsa's nodes and whatever machine he is peering with. Pcap/winpcap is a possibility based on some experience I have had with winpcap. But it would be good to hear if there is anything else that is unusual in his setup that might cause this. I am guessing that there are many other nodes on HECnet that are also SIMH, if there are, they do not bounce nearly as often.
Regards
Rob
-----Original Message-----
From: owner-hecnet at Update.UU.SE [mailto:owner-hecnet at Update.UU.SE] On
Behalf Of Gregg Levine
Sent: 29 June 2014 23:50
To: hecnet at update.uu.se
Subject: Re: [HECnet] Adjacencies Keep Bouncing
Hello!
Now this is straight out into deep space, and probably aiming for a cloaked
starship, but what if the problem is how the network is connected to what
Sampsa runs?
Sampsa how about some input from you please?
-----
Gregg C Levine gregg.drwho8 at gmail.com
"This signature fought the Time Wars, time and again."
On Sun, Jun 29, 2014 at 12:38 PM, Johnny Billquist <bqt at softjar.se> wrote:
Just a few comments...
8.400 (GORVAX) is a SIMH VAX, according to the nodename database.
http://madame.update.uu.se/~bqt/nodedb?search=gorvax&field=0&sort=0
I see the same issues with GORVAX and KUHAVX (47.556) also on MIM.
Both machines (and areas) are run by Sampsa. Not sure how relevant
that is...
I have not done any further looking at actual packets yet. I might if
I find some time. But it do seem there is some kind of problem/issue here.
Johnny
On 2014-06-29 02:04, Robert Jarratt wrote:
I think this was discussed not too long ago, but I can t find the
email thread. Sorry, this is quite a long email.
Here at 5.1023 I keep experiencing certain adjacencies going up and
down. At the bottom of this email is an extract of several hours of
my log (all times are UK local time, current GMT+1).
Last night I investigated one of these, when 8.400 went down at
2014-06-28 22:04:03. I used a packet sniffer and could see that my
Ethernet Router Hello message went out, with 8.400 in it, at the
following times:
22:03:18
22:03:23
22:03:33
22:03:38
22:03:48
22:03:53
I did not send any Ethernet Router Hello messages in that period that
excluded 8.400.
Conversely, from 8.400 I received Ethernet Router Hello messages,
with
5.1023 included except where noted at the following times:
22:03:28
22:03:29
22:03:30
22:03:45
22:03:46
22:03:47
22:04:03 (5.1023 missing)
22:04:04
22:04:07
22:04:08
22:04:09
22:04:25
22:04:26
So here are some interesting things about the above information.
1. I am supposed to send out the Ethernet Router Hello every 15 seconds.
Hello!
Now this is straight out into deep space, and probably aiming for a
cloaked starship, but what if the problem is how the network is
connected to what Sampsa runs?
Sampsa how about some input from you please?
-----
Gregg C Levine gregg.drwho8 at gmail.com
"This signature fought the Time Wars, time and again."
On Sun, Jun 29, 2014 at 12:38 PM, Johnny Billquist <bqt at softjar.se> wrote:
Just a few comments...
8.400 (GORVAX) is a SIMH VAX, according to the nodename database.
http://madame.update.uu.se/~bqt/nodedb?search=gorvax&field=0&sort=0
I see the same issues with GORVAX and KUHAVX (47.556) also on MIM.
Both machines (and areas) are run by Sampsa. Not sure how relevant that
is...
I have not done any further looking at actual packets yet. I might if I find
some time. But it do seem there is some kind of problem/issue here.
Johnny
On 2014-06-29 02:04, Robert Jarratt wrote:
I think this was discussed not too long ago, but I can t find the email
thread. Sorry, this is quite a long email.
Here at 5.1023 I keep experiencing certain adjacencies going up and
down. At the bottom of this email is an extract of several hours of my
log (all times are UK local time, current GMT+1).
Last night I investigated one of these, when 8.400 went down at
2014-06-28 22:04:03. I used a packet sniffer and could see that my
Ethernet Router Hello message went out, with 8.400 in it, at the
following times:
22:03:18
22:03:23
22:03:33
22:03:38
22:03:48
22:03:53
I did not send any Ethernet Router Hello messages in that period that
excluded 8.400.
Conversely, from 8.400 I received Ethernet Router Hello messages, with
5.1023 included except where noted at the following times:
22:03:28
22:03:29
22:03:30
22:03:45
22:03:46
22:03:47
22:04:03 (5.1023 missing)
22:04:04
22:04:07
22:04:08
22:04:09
22:04:25
22:04:26
So here are some interesting things about the above information.
1. I am supposed to send out the Ethernet Router Hello every 15 seconds.
Just a few comments...
8.400 (GORVAX) is a SIMH VAX, according to the nodename database.
http://madame.update.uu.se/~bqt/nodedb?search=gorvax&field=0&sort=0
I see the same issues with GORVAX and KUHAVX (47.556) also on MIM.
Both machines (and areas) are run by Sampsa. Not sure how relevant that is...
I have not done any further looking at actual packets yet. I might if I find some time. But it do seem there is some kind of problem/issue here.
Johnny
On 2014-06-29 02:04, Robert Jarratt wrote:
I think this was discussed not too long ago, but I can t find the email
thread. Sorry, this is quite a long email.
Here at 5.1023 I keep experiencing certain adjacencies going up and
down. At the bottom of this email is an extract of several hours of my
log (all times are UK local time, current GMT+1).
Last night I investigated one of these, when 8.400 went down at
2014-06-28 22:04:03. I used a packet sniffer and could see that my
Ethernet Router Hello message went out, with 8.400 in it, at the
following times:
22:03:18
22:03:23
22:03:33
22:03:38
22:03:48
22:03:53
I did not send any Ethernet Router Hello messages in that period that
excluded 8.400.
Conversely, from 8.400 I received Ethernet Router Hello messages, with
5.1023 included except where noted at the following times:
22:03:28
22:03:29
22:03:30
22:03:45
22:03:46
22:03:47
22:04:03 (5.1023 missing)
22:04:04
22:04:07
22:04:08
22:04:09
22:04:25
22:04:26
So here are some interesting things about the above information.
1. I am supposed to send out the Ethernet Router Hello every 15 seconds.
I think this was discussed not too long ago, but I can t find the email thread. Sorry, this is quite a long email.
Here at 5.1023 I keep experiencing certain adjacencies going up and down. At the bottom of this email is an extract of several hours of my log (all times are UK local time, current GMT+1).
Last night I investigated one of these, when 8.400 went down at 2014-06-28 22:04:03. I used a packet sniffer and could see that my Ethernet Router Hello message went out, with 8.400 in it, at the following times:
22:03:18
22:03:23
22:03:33
22:03:38
22:03:48
22:03:53
I did not send any Ethernet Router Hello messages in that period that excluded 8.400.
Conversely, from 8.400 I received Ethernet Router Hello messages, with 5.1023 included except where noted at the following times:
22:03:28
22:03:29
22:03:30
22:03:45
22:03:46
22:03:47
22:04:03 (5.1023 missing)
22:04:04
22:04:07
22:04:08
22:04:09
22:04:25
22:04:26
So here are some interesting things about the above information.
1. I am supposed to send out the Ethernet Router Hello every 15 seconds. From my router log, that is exactly what happens (at :18, :33, :48 etc). However the packet sniffer sees extra packets inserted. I have no idea where these packets are coming from (although my suspicion is winpcap, see later).
2. 8.400 seems to send its Ethernet Router Hello every 15 seconds, but repeats them several times at 1-second intervals. In the extract above it is repeated 3 times, then 3 times, then 5 times.
3. For a node to decide to drop an adjacency (ie to exclude it from the Ethernet Router Hello message), it must fail to see an Ethernet Router Hello for 3xHello Timer (Hello Timer is 15 seconds by default, so 45 seconds). Clearly my side has sent several Ethernet Router Hello messages in that time. If this was UDP dropping packets, then I would have hoped that it would not drop my particular ones 3 times in a row over a period of 45 seconds.
So, somehow, my Ethernet Router Hello messages are being dropped on their way to 8.400. However, this can t be outbound from 5.1023, because if that was the case then all my adjacencies would all go down at the same time. At any one time, only adjacency goes down. This suggests that the packet is somehow getting lost towards the end of its journey to 8.400.
What kind of node is 8.400? 47.556 is another one that appears to have this problem, I see that one is SIMH.
I am wondering if pcap/winpcap might be implicated here. The reason is that I noticed winpcap seems to drop incoming packets if you have sent a burst of outbound packets. In my router I was sending all the DECnet Level 1 Routing messages, all at the same time and getting some problems similar to this with a node local to my network, because my router did not see the packets coming from the other node, even though the sniffer could see them. I changed my router code to send the Level 1 Routing messages with a 1-second delay between them, and that problem went away. So, I am wondering if the remote nodes I have this problem with are SIMH nodes?
I do also see, less frequently, timeouts on my side. These are when I fail to see an Ethernet Router Hello from the other node for 45 seconds. I have not analysed this scenario.
Any ideas?
Regards
Rob
2014-06-28 15:00:50 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
2014-06-28 15:42:04 Adjacency timeout 8.400 (Slot 23)
2014-06-28 15:42:04 Adjacency down 8.400 (Slot 23)
2014-06-28 15:42:06 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
2014-06-28 16:37:12 Adjacency timeout 8.400 (Slot 23)
2014-06-28 16:37:12 Adjacency down 8.400 (Slot 23)
2014-06-28 16:37:17 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
2014-06-28 18:11:46 Adjacency timeout 8.400 (Slot 23)
2014-06-28 18:11:46 Adjacency down 8.400 (Slot 23)
2014-06-28 18:12:04 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
2014-06-28 18:15:11 Adjacency timeout 47.556 (Slot 25)
2014-06-28 18:15:11 Adjacency down 47.556 (Slot 25)
2014-06-28 18:15:14 Adjacency up 47.556 (Slot 25) on psilo.update.uu.se
2014-06-28 18:33:15 Adjacency timeout 47.556 (Slot 25)
2014-06-28 18:33:15 Adjacency down 47.556 (Slot 25)
2014-06-28 18:33:18 Adjacency up 47.556 (Slot 25) on psilo.update.uu.se
2014-06-28 18:33:19 Adjacency down 47.556 (Slot 25)
2014-06-28 18:33:42 Adjacency up 47.556 (Slot 25) on psilo.update.uu.se
2014-06-28 18:54:37 Adjacency timeout 47.556 (Slot 25)
2014-06-28 18:54:37 Adjacency down 47.556 (Slot 25)
2014-06-28 18:54:43 Adjacency up 47.556 (Slot 25) on psilo.update.uu.se
2014-06-28 19:36:02 Adjacency timeout 47.556 (Slot 25)
2014-06-28 19:36:02 Adjacency down 47.556 (Slot 25)
2014-06-28 19:36:20 Adjacency up 47.556 (Slot 25) on psilo.update.uu.se
2014-06-28 19:44:36 Adjacency timeout 28.41 (Slot 24)
2014-06-28 19:44:36 Adjacency down 28.41 (Slot 24)
2014-06-28 19:53:04 Adjacency up 28.41 (Slot 24) on psilo.update.uu.se
2014-06-28 19:55:08 Adjacency timeout 28.41 (Slot 24)
2014-06-28 19:55:08 Adjacency down 28.41 (Slot 24)
2014-06-28 19:57:20 Adjacency up 28.41 (Slot 24) on psilo.update.uu.se
2014-06-28 20:00:03 Adjacency down 8.400 (Slot 23)
2014-06-28 20:00:03 Adjacency down 47.556 (Slot 25)
2014-06-28 20:00:04 Adjacency up 47.556 (Slot 25) on psilo.update.uu.se
2014-06-28 20:00:05 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
2014-06-28 20:18:08 Adjacency timeout 8.400 (Slot 23)
2014-06-28 20:18:08 Adjacency down 8.400 (Slot 23)
2014-06-28 20:18:12 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
2014-06-28 20:47:46 Adjacency down 8.400 (Slot 23)
2014-06-28 20:47:46 Adjacency down 47.556 (Slot 25)
2014-06-28 20:47:49 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
2014-06-28 20:47:49 Adjacency up 47.556 (Slot 25) on psilo.update.uu.se
2014-06-28 22:04:03 Adjacency down 8.400 (Slot 23)
2014-06-28 22:04:04 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
2014-06-28 23:10:26 Adjacency timeout 5.99 (Slot 30)
2014-06-28 23:10:26 Adjacency down 5.99 (Slot 30)
2014-06-28 23:18:28 Adjacency down 8.400 (Slot 23)
2014-06-28 23:18:28 Adjacency down 47.556 (Slot 25)
2014-06-28 23:18:34 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
2014-06-28 23:18:35 Adjacency up 47.556 (Slot 25) on psilo.update.uu.se
2014-06-28 23:25:50 Adjacency timeout 47.556 (Slot 25)
2014-06-28 23:25:50 Adjacency down 47.556 (Slot 25)
2014-06-28 23:25:54 Adjacency up 47.556 (Slot 25) on psilo.update.uu.se
2014-06-29 00:06:52 Adjacency timeout 8.400 (Slot 23)
2014-06-29 00:06:52 Adjacency down 8.400 (Slot 23)
2014-06-29 00:06:58 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
2014-06-29 00:10:47 Adjacency timeout 47.556 (Slot 25)
2014-06-29 00:10:47 Adjacency down 47.556 (Slot 25)
2014-06-29 00:10:47 Adjacency up 47.556 (Slot 25) on psilo.update.uu.se
2014-06-29 00:28:21 Adjacency timeout 8.400 (Slot 23)
2014-06-29 00:28:21 Adjacency down 8.400 (Slot 23)
2014-06-29 00:28:26 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
2014-06-29 00:30:54 Adjacency down 8.400 (Slot 23)
2014-06-29 00:31:18 Adjacency timeout 47.556 (Slot 25)
2014-06-29 00:31:18 Adjacency down 47.556 (Slot 25)
2014-06-29 00:31:21 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
2014-06-29 00:31:21 Adjacency up 47.556 (Slot 25) on psilo.update.uu.se
2014-06-29 00:45:41 Adjacency timeout 8.400 (Slot 23)
2014-06-29 00:45:41 Adjacency down 8.400 (Slot 23)
2014-06-29 00:46:04 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
2014-06-29 00:53:57 Adjacency down 11.2 (Slot 21)
2014-06-29 00:54:05 Adjacency up 11.2 (Slot 21) on psilo.update.uu.se
2014-06-29 04:45:36 Adjacency timeout 47.556 (Slot 25)
2014-06-29 04:45:36 Adjacency down 47.556 (Slot 25)
2014-06-29 04:45:38 Adjacency up 47.556 (Slot 25) on psilo.update.uu.se
2014-06-29 04:49:49 Adjacency timeout 47.556 (Slot 25)
2014-06-29 04:49:49 Adjacency down 47.556 (Slot 25)
2014-06-29 04:53:19 Adjacency up 47.556 (Slot 25) on psilo.update.uu.se
2014-06-29 06:46:04 Adjacency up 62.637 (Slot 29) on psilo.update.uu.se
2014-06-29 08:01:32 Adjacency timeout 47.556 (Slot 25)
2014-06-29 08:01:32 Adjacency down 47.556 (Slot 25)
2014-06-29 08:01:33 Adjacency timeout 8.400 (Slot 23)
2014-06-29 08:01:33 Adjacency down 8.400 (Slot 23)
2014-06-29 08:03:04 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
2014-06-29 08:03:51 Adjacency timeout 8.400 (Slot 23)
2014-06-29 08:03:51 Adjacency down 8.400 (Slot 23)
2014-06-29 08:04:39 Adjacency timeout 47.556 (Slot 23)
2014-06-29 08:04:49 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
2014-06-29 08:04:50 Adjacency up 47.556 (Slot 25) on psilo.update.uu.se
2014-06-29 08:05:52 Adjacency timeout 47.556 (Slot 25)
2014-06-29 08:05:52 Adjacency down 47.556 (Slot 25)
2014-06-29 08:08:06 Adjacency up 47.556 (Slot 25) on psilo.update.uu.se
2014-06-29 08:10:02 Adjacency timeout 8.400 (Slot 23)
2014-06-29 08:10:02 Adjacency down 8.400 (Slot 23)
2014-06-29 08:10:18 Adjacency up 8.400 (Slot 23) on psilo.update.uu.se
Cory,
If you want to share the SYSUAF.DAT (and you technically don't have to in a VMS cluster, but it gets a bit hairy with multiple ones), you would put it on a common disk directly mounted by all nodes and pointed to by the SYSUAF system logical name, oh and usually a bunch of other files are included with this, There is a template command procedure in SYS$STARTUP that describes them, I believe SYLOGICALS.
A quorum disk is only needed to provide a vote, and usually only used in a two node cluster.
Recent VMS versions do have external authentication capabilities but you still have a SYSUAF.DAT as a backup...
- Dave
On 6/12/2014 5:40 PM, Cory Smelosky wrote:
Got a (hopefully) quick question regarding clustering
I've got a VAX<---->Alpha cluster rolling (VAX for booting VAX
satellites, Alpha for Alpha satellites) and I'm running in to an issue
with the SYS$UAF stuff. Only the password/user DB...nothing else needs
shared.
What's the easiest way to share it between the VAX and Alpha nodes
WITHOUT a using a quorum disk?
1). Kerberos or something? (I could have both query active directory...)
2). Shove SYS$UAF on MOIRA$dka100, mount in SYLOGICALS on all nodes?
3). Other
Got a (hopefully) quick question regarding clustering
I've got a VAX<---->Alpha cluster rolling (VAX for booting VAX satellites, Alpha for Alpha satellites) and I'm running in to an issue with the SYS$UAF stuff. Only the password/user DB...nothing else needs shared.
What's the easiest way to share it between the VAX and Alpha nodes WITHOUT a using a quorum disk?
1). Kerberos or something? (I could have both query active directory...)
2). Shove SYS$UAF on MOIRA$dka100, mount in SYLOGICALS on all nodes?
3). Other
--
Cory Smelosky
http://gewt.net Personal stuff
http://gimme-sympathy.org Projects
For those working on parallel computing: rullfs::[.parallel]
contains some routines in Fortran for VMS, inspired by TCGMSG
and the paper with the SEM routines.
Erik
I think I might just wipe the drive and install from scratch. It has gotten a bit full of cruft over time. ;)
Sent from my mobile^Wcorporate advertising device
On 11 Jun 2014, at 05:40, Cory Smelosky <b4 at gewt.net> wrote:
On Wed, 11 Jun 2014, Hans Vlems wrote:
What is the value for intstackpages? May be as low as 4 and must be at least 6 (iirc)
25.
Also: progress:
MOIRA::CSMELOSKY$ type sys$manager:DECW$SERVER_0_ERROR.LOG
11-JUN-2014 05:35:07.0 Hello, this is the X server
This is the DECwindows X11 display server for OpenVMS VAX V7.1-010215
compiled on Feb 15 2001 at 08:53:03
Dixmain address=00019c58
Server is running in bug-compatible mode
Now attach all known txport images
Can't start transports
Yet:
$ decw$device == "GAA0,GAB0,WSA1,WSA2,WSA3,WSA4,WSA4"
$ ! decw$default_keyboard_map == "US_LK401AA"
$ decw$server_transports == "DECNET,LOCAL,TCPIP"
--
Cory Smelosky
http://gewt.net Personal stuff
http://gimme-sympathy.org Projects
On Wed, 11 Jun 2014, Hans Vlems wrote:
What is the value for intstackpages? May be as low as 4 and must be at least 6 (iirc)
25.
Also: progress:
MOIRA::CSMELOSKY$ type sys$manager:DECW$SERVER_0_ERROR.LOG
11-JUN-2014 05:35:07.0 Hello, this is the X server
This is the DECwindows X11 display server for OpenVMS VAX V7.1-010215
compiled on Feb 15 2001 at 08:53:03
Dixmain address=00019c58
Server is running in bug-compatible mode
Now attach all known txport images
Can't start transports
Yet:
$ decw$device == "GAA0,GAB0,WSA1,WSA2,WSA3,WSA4,WSA4"
$ ! decw$default_keyboard_map == "US_LK401AA"
$ decw$server_transports == "DECNET,LOCAL,TCPIP"
--
Cory Smelosky
http://gewt.net Personal stuff
http://gimme-sympathy.org Projects
What is the value for intstackpages? May be as low as 4 and must be at least 6 (iirc)
Verzonden vanaf mijn BlackBerry 10-smartphone.
Origineel bericht
Van: Brian Schenkenberger, VAXman-
Verzonden: woensdag 11 juni 2014 03:12
Aan: hecnet at Update.UU.SE
Beantwoorden: hecnet at Update.UU.SE
Onderwerp: Re: [HECnet] Latest version of VWS
Cory Smelosky <b4 at gewt.net> writes:
On Tue, 10 Jun 2014, Brian Schenkenberger, VAXman- wrote:
8192. Looks like SET FILE/ATTRIB=LRL:8192 EWS012.A allowed BACKUP/LIST
EWS012.A/SAVE to succeed.
VMS DUMP the first block. You should see the BACKUP command in the ASCII
translation. If there's a /BLOCK listed, use that value; otherwise, there
are defaults as you have discovered.
Didn't even need to DUMP it. VMSINSTAL didn't complain.
Don't feel bad, I've been dealing with this for 20 years when trying to get
VMS crash dump files from customers sent to me in tact via the internet. ;)
I'm impressed I figured out that LRL was the value I wanted. ;)
Luck. ;)
Yup. ;)
Next battle: anything graphical logs this:
%%%%%%%%%%% OPCOM 10-JUN-2014 19:25:19.33 %%%%%%%%%%%
Message from user AUDIT$SERVER on MOIRA
Security alarm (SECURITY) and security audit (SECURITY) on MOIRA, system
id: 9217
Auditable event: Detached process login failure
Event time: 10-JUN-2014 19:25:19.32
PID: 20200149
Process name: _WSA3:
Username: SYSTEM
Process owner: [SYSTEM]
Image name: MOIRA$DKA0:[SYS0.SYSCOMMON.][SYSEXE]LOGINOUT.EXE
Status: %DECW-F-NOMSG, Message number 02DB8204
Which makes no sense.
$ SET MESSAGE SYS$MESSAGE:DECW$XLIBMSG.EXE
$ EXIT %X02DB8204
%XLIB-F-CANTOPEN, cannot open display
Do you, indeed, have the DECWindows server started?
Is there a Gxx0: device? Configured? Channels assigned?
Try setting SYSGEN parameter STARTUP_P2 to "VDC" in a conversational boot
and then continue. You will find a STARTUP.LOG file in SYS$SYSTEM which
might hold some clues. ;)
--
VAXman- A Bored Certified VMS Kernel Mode Hacker VAXman(at)TMESIS(dot)ORG
I speak to machines with the voice of humanity.
