I'm receiving some dupes that shouldn't happen. For echomail messages
that originate or go through systems using hpt 1.9, which have
-sbkeepall in their config set.
For instance here's a routing diagram of what I see for messages
generated on 3:770/1 in the STATS area:
+----------+ A +-----------+
+ 3:770/1 |------->| 3:712/848 |
+----------+ +-----------+
| |
B | +-----------+ | C
+-->| 2:280/464 |<--+
+-----------+
A: Only Z3 downlinks of 3:770/1 in the seenby's
B: All downlinks (in all zones) of 3:770/1 in the seenby's
C: Unnecessary dupes received through 3:712/848, because my node
should be in the seen+by's of the message received by 3:712/848 from 3:770/1.
An example for today:
From : 3:770/1 2024-01-31 23:59:16
To : All
Subj : Daily Areas Summary
@MSGID: 3:770/1 65ba2805
Has the following seen+by's and path when received directly from
3:770/1 by my system:
SEEN+BY: 218/840 220/70 226/17 267/800 280/464 712/848 770/1 100 330
340 SEEN+BY: 772/220 230
PATH: 770/1
And these when received through 3:712/848 as dupe:
SEEN+BY: 229/426 280/464 301/1 633/267 712/620 848 770/1 100 330 340 772/220 SEEN+BY: 772/230
PATH: 770/1 712/848
Notice that all the Z1 downlinks of 3:770/1 (218/840 220/70 226/17 267/800) present in the direct message, are missing from the seen+by's
in the message that is received as dupe from 3:712/848.
I have confirmation it's the originating hpt system where this is happening, and there is no seen+by stripping taking place on the
system in between.
3:770/1 Seems to be using hpt/lnx 1.9 2022-07-03
I've also confirmed the same thing happening on another system using hpt/lnx 1.9 2023-02-24.
It is more important to know whether 3:712/848 uses hpt.
I cannot promise I'll be able to address this issue any time soon, but to do it I must have the hpt configuration from 3:712/848 and two messages: the message he receives from 3:770/1 and the message he sends to you. My email address is in the tagline.
3:770/1 Seems to be using hpt/lnx 1.9 2022-07-03
It is more important to know whether 3:712/848 uses hpt.
I've also confirmed the same thing happening on another system using
hpt/lnx 1.9 2023-02-24.
I cannot promise I'll be able to address this issue any time soon, but to do it I must have the hpt configuration from 3:712/848 and two messages: the message he receives from 3:770/1 and the message he sends to you. My email address is in the tagline.
I have attached two packets (zipped) for you via email that were
created here by hpt. One packet was destined to a Z1 link, the other
was destined to a Z2 link. This is before they even left my system, so obviously unaltered by another.
If you need anything else, please let me know.
3:770/1 Seems to be using hpt/lnx 1.9 2022-07-03
It is more important to know whether 3:712/848 uses hpt.
He does, but it's not relevant.
I've also seen this happen where the second system doesn't use hpt.
This involves:
1:154/10 hpt
1:154/30 synchronet
2:280/464 me (using FMail)
Where 1:154/10 also posts messages in the STATS area. And I'm getting
them also as dupes from 1:154/30.
1:154/10 Investigated his outbound .pkt files for me and 1:154/30, the seen+by's where different, where they should be the same.
I've also confirmed the same thing happening on another system
using hpt/lnx 1.9 2023-02-24.
I cannot promise I'll be able to address this issue any time
soon, but to do it I must have the hpt configuration from
3:712/848 and two messages: the message he receives from 3:770/1
and the message he sends to you. My email address is in the
tagline.
Maybe I can give you those for the 1:154/* case above. But it really
is as I told you, it was thoroughly investigated!
But it should be easy for you to test it yourself. Just write a
message in an echomail area for which you have links in multiple
zones, and check the seen+by's in the generated outbound .pkt files
for these links. They should all be the same, but they won't be...
I'm receiving some dupes that shouldn't happen. For echomail messages
that originate or go through systems using hpt 1.9, which have
-sbkeepall in their config set.
I'm receiving some dupes that shouldn't happen. For echomail messages
that originate or go through systems using hpt 1.9, which have
-sbkeepall in their config set.
I've pushed a fix. Please test it.
But if my links, that use husky, start using this version, I will keep a close eye on what will happen!
I've pushed a fix. Please test it.
^^But if my links, that use husky, start using this version, I will
keep a close eye on what will happen!
Updated. Let me know what you see from here on out.
Regards,
Nick
SEEN-BY: 103/705 120/616 123/10 124/5016 153/757 154/10 30 40 50 700
@PATH: 154/10 280/464
^^
I didn't also get this message as dupe from 154/30, which is connected here
aswell. So the first "test" seems to be successful! ;-)
I've pushed a fix. Please test it.
But if my links, that use husky, start using this version, I will keep
a close eye on what will happen!
I've pushed a fix. Please test it.
I didn't also get this message as dupe from 154/30, which is
connected here aswell. So the first "test" seems to be successful!
;-)
Did you only quote one line of the SEEN+BY?
For some wierd reason it seems like there should be more on that list.
;)
Otherwise, let me know how the test goes!
@TID: hpt/lnx 1.9 2024-02-05
But if my links, that use husky, start using this version, I will
keep a close eye on what will happen!
Updated here...
It looks good! I didn't receive any dupes for this message from you. Neither from 712/848 or 770/1, that are both directly connected here
to this area. But they should also upgrade!
On Monday February 05 2024, Michael Dukelsky said the following...
I've pushed a fix. Please test it.
Thanks Michael,
I've updated here. I have automated messages I export to zones 1, 2 &
3 each morning in the horoscope echo, I'll check on these tomorrow &
see what they look like.
* Carbon copied to Paul Hayton 3:770/1
Hi Stephen,
On 2024-02-06 09:06:29, I wrote to you:
It looks good! I didn't receive any dupes for this message from
you. Neither from 712/848 or 770/1, that are both directly
connected here to this area. But they should also upgrade!
Paul Hayton doesn't seem to have updated his hpt yet. I still got his messages in the STATS area as dupes through 712/848...
way to do it? I'll try to find some time to have a look at this in the coming week.
It's been a while since I built this software and looking back at my notes I used to do this with success
I would do a fresh install in a new husky directory. I just did that a
few days ago and all is well here.
From section 2.4 of the INSTALL.asciidoc ..
wget https://raw.githubusercontent.com/huskyproject/huskybse/master/script/init
chmod 0755 init_build
./init_build
Then change into the husky directory and adjust the huskymac.cfg as
needed and then run..
./build.sh
OK all done, seemed to work OK but am not 100% sure if I have the new versions running (or not)?
Wilfred do you want to run another test to see if 3:770/1 is behaving itself now?
Hopefully I'm now in sync with the rest of the community :)
Hopefully I'm now in sync with the rest of the community :)
Just run 'hpt' on the command line with no arguments. First line
before the usage help will tell you.
'hpt/lnx 1.9 [date here]' -- 2024-02-05 is what you're looking for.
If not, you did it wrong. :)
OK all done, seemed to work OK but am not 100% sure if I have the new versions running (or not)?
Wilfred do you want to run another test to see if 3:770/1 is behaving itself now?
OK all done, seemed to work OK but am not 100% sure if I have the
new versions running (or not)?
Wilfred do you want to run another test to see if 3:770/1 is behaving
itself now?
When your daily messages in the STATS area are posted I'll know when there
are no dupes coming in through Scott.
There is no "-v" in hpt. However, it is there in htick. :)
On Sat, 17 Feb 2024 14:36:52 +0200, Tommi Koivula -> Nicholas Boel wrote:
There is no "-v" in hpt. However, it is there in htick. :)Yes, I know. That is why I didn't tell him to use '-v'. Just executing 'hpt' with no command line arguments will give him the same result, including a help screen! :)
Yep, my message was a hint to the developers. ;-)
Just run 'hpt' on the command line with no arguments. First line before the usage help will tell you.
'hpt/lnx 1.9 [date here]' -- 2024-02-05 is what you're looking for. If not, you did it wrong. :)
When your daily messages in the STATS area are posted I'll know when
there are no dupes coming in through Scott.
Those were posted about 21 minutes ago, and there are no dupes sofar. So it's looking good!
Sysop: | Angel Ripoll |
---|---|
Location: | Madrid, Spain |
Users: | 11 |
Nodes: | 8 (0 / 8) |
Uptime: | 36:19:36 |
Calls: | 479 |
Files: | 14,069 |
Messages: | 62,132 |