View Issue Details

IDProjectCategoryView StatusLast Update
0004542secusharesocialpublic2019-02-12 09:13
ReporterlynXAssigned To 
PrioritynormalSeverityminorReproducibilityalways
Status confirmedResolutionopen 
Product Version 
Target VersionFixed in Version 
Summary0004542: pubsub (gnunet-social) does not work with separated system/user configs
DescriptionUpdate: This bug describes how gnunet-social is unable to relay packets to its lower layers which are running on another uid as intentional with the SYSTEM_ONLY + USER_ONLY style of configuration.

----

I wrote the "psycit" script to ensure gnunet is running in a sane
configuration and suitable egos have been generated. It prototypes
the kind of jobs that a secushare application would also have to
automate. Then I ran it on one userid to host a chatroom. It would
fetch the previously generated ego and ask gnunet-social to produce
a chatroom.

computer] ./psycit -C
=> gnunet-social -e 0A9RPQ1VDZB3M67XEBP9MVF6ZHKA4TP3YS99FMVG9JF4TQ3C9SXG --follow --host-enter --welcome
May 29 17:30:56-912320 gnunet-social-26050 WARNING Ego: 0A9RPQ1VDZB3M67XEBP9MVF6ZHKA4TP3YS99FMVG9JF4TQ3C9SXG vonlynX
May 29 17:30:56-912862 gnunet-social-26050 WARNING Ego: X33YY4YG62PQVN6918TX812ZT4EJYZAGEJDJ6STNTS2Z3HEX8WVG lynX
May 29 17:30:56-913011 gnunet-social-26050 WARNING Ego: KXM3ER7Z8J50AKFNHJ31APRKVKYPSPPA37D4SN0G7A381N4EVYE0 user
May 29 17:30:56-913174 gnunet-social-26050 WARNING Host: HXWCGDE3WBFNBE0FMFNVGMQGB3480698AHQ5Z2EEZEV6BFJ1KES0
May 29 17:30:56-913328 gnunet-social-26050 WARNING Host: GGSE8Y2RVDWM9G2W02FHJKTXC8K3MDY8Z25V81EAXNWNE9329B10

Interestingly, gnunet-social lists also previously generated
egos although we explicitly didn't want to use those, then it
outputs a new (GGSE) and an old chatroom keydress (HXWC).
After that it waits for events ("--follow").

Next I switch to a different userid on the same machine
and pass it the keydress of the newly generated place. As you
can see it quietly pulls out its own ego from gnunet-identity,
then requests a --guest-enter with TVV5 being the locally
running node.

computer ~% psycit -p GGSE8Y2RVDWM9G2W02FHJKTXC8K3MDY8Z25V81EAXNWNE9329B10 -S
=> gnunet-social --guest-enter --follow -e H85F4CKSGWF41HCFNR97E9HH6CM24P46CTS64MQQ33Z0QXMCA2K0 -p GGSE8Y2RVDWM9G2W02FHJKTXC8K3MDY8Z25V81EAXNWNE9329B10 -i TVV54YYR29KRDKM30ZC3014JXHDGPVT7KFQGA44CEV2YQSEVFDAG
May 29 17:34:29-818344 gnunet-social-26127 WARNING Ego: H85F4CKSGWF41HCFNR97E9HH6CM24P46CTS64MQQ33Z0QXMCA2K0 egal
May 29 17:34:29-818975 gnunet-social-26127 WARNING Guest: HXWCGDE3WBFNBE0FMFNVGMQGB3480698AHQ5Z2EEZEV6BFJ1KES0

The debug warnings show HXWC which probably got there by
a previous test. Next up, nothing happens. Typing anything
into stdin of any of the two processes has no effect.
Submitting a --guest-talk has no visible effects, either:

computer ~% psycit -p GGSE8Y2RVDWM9G2W02FHJKTXC8K3MDY8Z25V81EAXNWNE9329B10 -P
=> gnunet-social --guest-talk -m _notice_testing_psycit -d 'check check' -e H85F4CKSGWF41HCFNR97E9HH6CM24P46CTS64MQQ33Z0QXMCA2K0 -p GGSE8Y2RVDWM9G2W02FHJKTXC8K3MDY8Z25V81EAXNWNE9329B10 -i TVV54YYR29KRDKM30ZC3014JXHDGPVT7KFQGA44CEV2YQSEVFDAG
May 29 17:34:29-818344 gnunet-social-26127 WARNING Ego: H85F4CKSGWF41HCFNR97E9HH6CM24P46CTS64MQQ33Z0QXMCA2K0 egal
May 29 17:34:29-818975 gnunet-social-26127 WARNING Guest: HXWCGDE3WBFNBE0FMFNVGMQGB3480698AHQ5Z2EEZEV6BFJ1KES0

Hitting ^C to terminate the experiment produces the following
assertion failure in both instances of gnunet-social:

^CMay 29 17:35:35-519194 gnunet-social-26050 ERROR Assertion failed at scheduler.c:978.
child died with signal 6, without coredump

^CMay 29 17:35:39-519965 gnunet-social-26127 ERROR Assertion failed at scheduler.c:978.
child died with signal 6, without coredump
TagsNo tags attached.

Activities

lynX

2016-05-30 07:49

manager  

psycit (4,150 bytes)
#!/usr/bin/perl
#
# use Net::PSYC;

sub debug() { 1 }

use Getopt::Std;
getopt('pime');

# pick the name of an ego from the environment
# FIXME: assuming nickname has no whitespace
my $nick = $opt_e
     || $ENV{'PSYCNICK'}
     || $ENV{'NICK'}		# this one should work with any chat system
     || $ENV{'IRCNICK'}
     || $ENV{'USER'}
     || 'user';

my $peerid = $opt_i
     || $ENV{'GNUNET_PEERID'}
	# peerid of the gnunet node runnning on cheet
     || 'TVV54YYR29KRDKM30ZC3014JXHDGPVT7KFQGA44CEV2YQSEVFDAG';

my $place = $opt_p
     || $ENV{'PSYCPLACE'}
	# pubkey of a place created on cheet
     || 'RG3RY7FTTT6128V1HGZQCQ89Q5APTHSP2JD9WKGH4T2N7KQZFVQG';

my $method = $opt_m || '_notice_testing_psycit';

# first, let's check if we are being run by the correct user
my $cf = "$ENV{HOME}/.config/gnunet.conf";
die "This user has no $cf that I can read" unless open(C, $cf);
my $ok = 0;
while(<C>) {
	# i'll check for [arm] some other day... FIXME
	$ok++ if /^\s*USER_ONLY\s*=\s*YES\s*$/i;
}
close C;
die "Your $cf is not configured to run user services" unless $ok;

# now, let's see if the services are up
my $cmd = 'gnunet-arm --info |';
die "$cmd failed: $!" unless open(C, $cmd);
$ok = 0;
while(<C>) {
	$ok++ if /^(gns|social|identity) /;
}
close C;
die "Not all of GNS, social and identity services are running" if $ok != 3;

unless ($ego = fetchEgo($nick)) {
	die "Could not create ego for $nick: $!"
	    unless sys('gnunet-identity', '--create', $nick);
	die "Could not fetch ego for $nick after creation: $!"
	    unless $ego = fetchEgo($nick);
	die "Could not assign $nick's ego as default for social: $!"
	    # FIXME: --set instead of --service doesnt look intuitive
	    unless sys('gnunet-identity', '--ego', $nick, '-s', 'social');
}



if ($opt_C) {
	# FIXME: gnunet-social should not need different commands here,
	#	 checking for presence of -p parameter should be enough.
	&sys( "gnunet-social -e $ego --follow ".( $opt_p? "--host-reconnect -p $opt_p"
							: "--host-enter --welcome" ));
} elsif ($opt_S) {
	# FIXME: why do we have differing API calls for entering and 
	#	 reconnecting a place?
	&sys( "gnunet-social --guest-enter --follow -e $ego -p $place -i $peerid" );
} elsif ($opt_P) {
	my $d = '';
	if ($#ARGV > 0) {
		$d = join(' ', @ARGV);
	} else {
		$d .= $_ while <STDIN>;
	}
	$d =~ tr/[\r\n]/ /;
	chomp $d;

	# FIXME: API must be able to know by itself if a place is local or remote
	#	 therefore the distinction of guest-talk and host-announce must go
	&sys( "gnunet-social --guest-talk -m $method -d '$d' -e $ego -p $place -i $peerid" );
} else {
	print <<X;

Hello, $nick! (aka $ego)

Usage: $0 <command> [-c] [-p <place>] [-i <peerid>]
		    [-m <method>] [-e <ego>] [ <body> ]

Commands are:
	-C	create or reconnect to a locally hosted pubsub channel
		(will not terminate as messages are expected)
	-S	enter or reconnect to a remotely hosted pubsub channel
		(will not terminate as messages are expected)
	-P	publish to a channel
	-h	show this help

Parameters are:
	-p		use this place where necessary
	-i		use this GNUnet PeerID where necessary
	-e		use this ego where necessary
	<body>		for publishing, can also be passed via stdin.

Options are:
	-c		continue in case of errors
	-m <method>	use a different method when publishing

X
#	-L	leave a channel
#	-s		show some status
}
exit;


sub sys {
	print "=> ", join(' ', @_), "\n" if debug;
	if (system(@_)) {
		if ($? == -1) {
			print "failed to execute: $!\n";
		} elsif ($? & 127) {
			printf "child died with signal %d, %s coredump\n",
			 ($? & 127),  ($? & 128) ? 'with' : 'without';
		} else {
			printf "child exited with value %d\n", $? >> 8;
		}
		exit $? unless $opt_c;
		return 0;
	}
	return 1;
}

sub fetchEgo {
	my $n = shift;
	# FIXME: there should be some UX consistency between gnunet tools
	#	 regarding --list, --info or --display
	# FIXME: gnunet-identity should offer a way to resolve exactly one
	#	 ego nickname
	my $e = undef, $cmd = 'gnunet-identity --display |';
	local(*C);
	die "$cmd failed: $!" unless open(C, $cmd);
	while(<C>) {
		# print;
		$e = $1 if /^$n - (\w+)$/i;	# may i ignore case?
	}
	close C;
	return $e;
}
psycit (4,150 bytes)

tg

2016-05-30 15:40

manager   ~0010779

the list you get at the beginning is the current status, it shows the current egos and entered places, same as --status (or no args)

--host-enter creates a new place, the address of which you can find on a line like this:
Host entered: <pubkey>

Then you can use that pubkey from other clients.

You were using one of the old place pubkeys instead of the one you just created, that's why it didn't work.

lynX

2016-05-30 16:50

manager   ~0010780

Well, then I guess the problem is that such a line "Host entered: <pubkey>" never appeared. What could be the cause for that?

tg

2016-05-30 17:10

manager   ~0010781

Last edited: 2016-05-30 18:25

View 2 revisions

make sure all of these services are running:
- social
- psyc
- psycstore
- multicast
- cadet

lynX

2016-05-30 18:27

manager   ~0010782

Oh wow, we've been running these tools for about a year now, and now we find out we should have started the services manually? That's amazing! How does one start such a service manually? Shouldn't they all AUTOSTART given such a statement is in each of their config files? Can you post a gnunet.conf or your complete set-up by which it works for you? Can you add code to all tools that detects when services are missing so users do not stay in the dark? Or even better, shouldn't all involved services be autostarted on demand?

tg

2016-05-30 18:34

manager   ~0010783

they all have AUTOSTART = YES by default,
but the question was: are all of these actually running on your system?

lynX

2016-05-30 18:43

manager   ~0010784

Well, I thought my answer made it clear that they are not. I just discovered that I can start them manually using 'gnunet-arm -i <service>'. That surprised me since last time I tried it didn't work. Okay, now I have all the services running. I run the same script as usual:

computer] ./psycit -C
=> gnunet-social -e 0A9RPQ1VDZB3M67XEBP9MVF6ZHKA4TP3YS99FMVG9JF4TQ3C9SXG --follow --host-enter --welcome
May 30 18:34:03-050696 gnunet-social-30564 WARNING Ego: 0A9RPQ1VDZB3M67XEBP9MVF6ZHKA4TP3YS99FMVG9JF4TQ3C9SXG vonlynX
May 30 18:34:03-051415 gnunet-social-30564 WARNING Ego: X33YY4YG62PQVN6918TX812ZT4EJYZAGEJDJ6STNTS2Z3HEX8WVG lynX
May 30 18:34:03-051731 gnunet-social-30564 WARNING Ego: KXM3ER7Z8J50AKFNHJ31APRKVKYPSPPA37D4SN0G7A381N4EVYE0 user
May 30 18:34:03-051976 gnunet-social-30564 WARNING Host: SQVPFMVCB3G1W5ACWPNV1XJGVYKXTMWWPMHYJA7VY73K5J9X7VF0
May 30 18:34:03-052296 gnunet-social-30564 WARNING Host: GGSE8Y2RVDWM9G2W02FHJKTXC8K3MDY8Z25V81EAXNWNE9329B10
May 30 18:34:03-052542 gnunet-social-30564 WARNING Host: JTWFK8BPAYK4NFV2F58WS0KSG7Y5FDF04HN8SB0J2TB5ZCTCH150
May 30 18:34:03-052740 gnunet-social-30564 WARNING Host: HXWCGDE3WBFNBE0FMFNVGMQGB3480698AHQ5Z2EEZEV6BFJ1KES0
May 30 18:34:03-052924 gnunet-social-30564 WARNING Host: Z9EXJC8ET45JYN8834AB4E5QQJARYRHY544N6KSA43GF4737YN60

It lists all the dead horses of the past, but still does not produce a "Host entered" message for the new place.

tg

2016-05-30 18:52

manager   ~0010785

Last edited: 2016-05-30 19:45

View 5 revisions

then checking the logs of services would help to figure out the problem

enable logging for all services:

[arm]
GLOBAL_POSTFIX = -l $GNUNET_CACHE_HOME/{}-%F.%T-[].log

then increase log level for specific services:

[multicast]
#AUTOSTART = NO
OPTIONS = -L DEBUG -l /home/gnunet/.cache/gnunet/multicast-%F.%T-[].log

[psycstore]
#AUTOSTART = NO
OPTIONS = -L DEBUG -l /home/gnunet/.cache/gnunet/social-%F.%T-[].log

[psyc]
#AUTOSTART = NO
OPTIONS = -L DEBUG -l /home/gnunet/.cache/gnunet/social-%F.%T-[].log

[social]
#AUTOSTART = NO
OPTIONS = -L DEBUG -l /home/gnunet/.cache/gnunet/social-%F.%T-[].log

(apparently $GNUNET_CACHE_HOME only works for GLOBAL_POSTFIX)

tg

2016-05-30 19:37

manager   ~0010788

Last edited: 2016-05-30 21:26

View 3 revisions

Also, for development work you can do the following.

First set AUTOSTART = NO in for each of these services,
then run each of the below commands in a separate terminal:
(gnunet-logread & gnunet-logread-ipc are in utils/)

Start IPC monitor:

rm -f /tmp/gnunet-logread-ipc.sock; gnunet-logread-ipc | gnunet-logread

Then start each of the services:

gnunet-service-multicast -L DEBUG |& tee multicast-$(date +%F.%T).log | gnunet-logread -L WARNING -i /tmp/gnunet-logread-ipc.sock -n multicast

gnunet-service-psycstore -L DEBUG |& tee psycstore-$(date +%F.%T).log | gnunet-logread -L WARNING -i /tmp/gnunet-logread-ipc.sock -n psycstore

gnunet-service-psyc -L DEBUG |& tee psyc-$(date +%F.%T).log | gnunet-logread -L WARNING -i /tmp/gnunet-logread-ipc.sock -n psyc

gnunet-service-social -L DEBUG |& tee social-$(date +%F.%T).log | gnunet-logread -L WARNING -i /tmp/gnunet-logread-ipc.sock -n social

tg

2016-05-30 19:40

manager   ~0010789

Last edited: 2016-05-30 19:49

View 2 revisions

just checked, the services start with socket activation,
i.e. when you run the gnunet-social client, the social service is started, which then causes all the rest to start when the services connect to each other

setting FORCESTART = YES skips socket activation and starts the service when you start arm

make sure all of these are running after starting gnunet-social,
then you can check the logs if there's any ERROR in there..
or just post them here

tg

2016-05-30 19:46

manager   ~0010790

Last edited: 2016-05-30 19:46

View 2 revisions

fixed the config in comment 10785

lynX

2016-05-30 21:05

manager   ~0010791

I removed some old egos with -D but apparently they were still checked into some places. Other than that, no progress:

computer] ./psycit -C
=> gnunet-social -e 0A9RPQ1VDZB3M67XEBP9MVF6ZHKA4TP3YS99FMVG9JF4TQ3C9SXG --follow --host-enter --welcome
May 30 20:25:22-168339 gnunet-social-10255 WARNING Ego: 0A9RPQ1VDZB3M67XEBP9MVF6ZHKA4TP3YS99FMVG9JF4TQ3C9SXG vonlynX
May 30 20:25:22-169220 gnunet-social-10255 WARNING Host: SQVPFMVCB3G1W5ACWPNV1XJGVYKXTMWWPMHYJA7VY73K5J9X7VF0
May 30 20:25:22-169497 gnunet-social-10255 ERROR Failure to obtain ego. Is identity service running?
May 30 20:25:22-169635 gnunet-social-10255 ERROR Assertion failed at social_api.c:962.
May 30 20:25:22-169832 gnunet-social-10255 WARNING Host: JTWFK8BPAYK4NFV2F58WS0KSG7Y5FDF04HN8SB0J2TB5ZCTCH150
May 30 20:25:22-170099 gnunet-social-10255 ERROR Failure to obtain ego. Is identity service running?
May 30 20:25:22-170263 gnunet-social-10255 ERROR Assertion failed at social_api.c:962.
May 30 20:25:22-170457 gnunet-social-10255 WARNING Host: Z9EXJC8ET45JYN8834AB4E5QQJARYRHY544N6KSA43GF4737YN60
May 30 20:25:22-170660 gnunet-social-10255 WARNING Host: VHWGYFWXAYFXVVP0A3NMCXT7XB0RTPMP11NMJXRNFYZE1HW6RXKG

With the new configuration two empty debug logs were created:

.cache/gnunet/gns-2016-05-30.log .cache/gnunet/namestore-2016-05-30.log

A more recent run made some debug output show up in the main gnunet.log.
Here's the tail of grep social- gnunet.log:

[skipped some adding to hashmap stuff]
May 30 20:45:35-755177 social-10706 DEBUG 0xb93d5cc8 Sending ego notification to client: vonlynX
May 30 20:45:35-771309 social-10706 DEBUG 0xb93d5cc8 Sending end of ego list notification to client
May 30 20:45:35-771432 social-10706 DEBUG 0xb93d5cc8 Sending place notification of type 841 to client.
May 30 20:45:35-771516 social-10706 DEBUG 0xb93d5cc8 Sending place notification of type 841 to client.
May 30 20:45:35-771598 social-10706 DEBUG 0xb93d5cc8 Sending place notification of type 841 to client.
May 30 20:45:35-771680 social-10706 DEBUG 0xb93d5cc8 Sending place notification of type 841 to client.
May 30 20:45:35-771762 social-10706 DEBUG 0xb93d5cc8 Sending place notification of type 841 to client.
May 30 20:45:35-771844 social-10706 DEBUG 0xb93d5cc8 Sending place notification of type 841 to client.
May 30 20:45:35-771926 social-10706 DEBUG 0xb93d5cc8 Sending place notification of type 841 to client.
May 30 20:45:35-772008 social-10706 DEBUG 0xb93d5cc8 Sending place notification of type 841 to client.
May 30 20:45:35-772090 social-10706 DEBUG 0xb93d5cc8 Sending end of place list notification to client
May 30 20:45:35-772199 social-10706 DEBUG 0xb93d0870 Application cli connected.
May 30 20:45:35-848574 social-10706 DEBUG Adding app place to hashmap:
May 30 20:45:35-848687 social-10706 DEBUG app_id = cli
May 30 20:45:35-848820 social-10706 DEBUG ego_place_pub_hash = GA38JNFC
May 30 20:45:35-848866 social-10706 DEBUG Adding place to hashmap:
May 30 20:45:35-848940 social-10706 DEBUG ego_place_pub_hash = GA38JNFC
May 30 20:45:35-849636 social-10706 WARNING 0xb93d76d0 Client connected as host to place M27K8JS4.
May 30 20:45:49-422860 social-10706 DEBUG 0xb93d76d0 Client (host) disconnected from place M27K8JS4

So it did do something, but the callback didn't call back.

lynX

2016-05-31 13:53

manager   ~0010810

After use of these gnunet-social commands, gnunet-arm -e hangs. Checking gnunet-arm -I shows that only "social" is still running. Since interrupting it with ^C it has gone into a non cooperative state. kill -9 gnunet helps. A simple kill is not enough. Here's what grep -v util- log shows:

May 31 13:12:03-049334 social-13445 DEBUG 0xb8784530 Sending place notification of type 841 to client.
May 31 13:12:03-049451 social-13445 DEBUG 0xb8784530 Sending place notification of type 841 to client.
May 31 13:12:03-049535 social-13445 DEBUG 0xb8784530 Sending place notification of type 841 to client.
May 31 13:12:03-049616 social-13445 DEBUG 0xb8784530 Sending end of place list notification to client
May 31 13:12:03-049698 social-13445 DEBUG 0xb8788a48 Application cli connected.
May 31 13:12:03-059293 social-13445 DEBUG Adding app place to hashmap:
May 31 13:12:03-059330 social-13445 DEBUG app_id = cli
May 31 13:12:03-059487 social-13445 DEBUG ego_place_pub_hash = 3THHXVFP
May 31 13:12:03-059529 social-13445 DEBUG Adding place to hashmap:
May 31 13:12:03-059602 social-13445 DEBUG ego_place_pub_hash = 3THHXVFP
May 31 13:12:03-060044 social-13445 DEBUG 0xb878a4a0 Client connected as guest to place QT5MNDW3.
May 31 13:14:53-707761 identity-13400 DEBUG Received START message from client
May 31 13:21:46-264807 social-13445 DEBUG 0xb878a4a0 Client (guest) disconnected from place QT5MNDW3
May 31 13:26:34-362512 identity-13400 DEBUG Received START message from client
May 31 13:28:35-958429 identity-13400 DEBUG Received START message from client
May 31 13:38:11-466917 gns-13398 DEBUG Shutting down!
May 31 13:38:11-467576 namestore-api-13398 DEBUG Cleaning up
May 31 13:38:11-467760 namecache-api-13398 DEBUG Cleaning up
May 31 13:38:11-470984 social-13445 DEBUG 0xb8787c40 Cleaning up place M27K8JS4
May 31 13:38:11-471293 social-13445 DEBUG 0xb8786668 Cleaning up place YJGS3YR7
May 31 13:38:11-471473 social-13445 DEBUG 0xb8788320 Cleaning up place 2PN8ENE4
May 31 13:38:11-471628 social-13445 DEBUG 0xb87857e0 Cleaning up place 1A5HP836
May 31 13:38:11-471785 social-13445 DEBUG 0xb87850b8 Cleaning up place 7BTKARX4
May 31 13:38:11-471947 social-13445 DEBUG 0xb8784a20 Cleaning up place BG6MP3BS
May 31 13:38:11-472402 social-13445 DEBUG 0xb8785ee8 Cleaning up place NVFCYP4N
May 31 13:38:11-472561 social-13445 DEBUG 0xb8786d38 Cleaning up place VRV0FFYN
May 31 13:38:11-472683 social-13445 DEBUG 0xb8787440 Cleaning up place KW0T3JCY
May 31 13:38:11-472804 social-13445 DEBUG 0xb878a4a0 Cleaning up place QT5MNDW3
May 31 13:38:11-472952 core-api-13445 DEBUG Disconnecting from CORE service
May 31 13:38:11-489555 namestore-api-13445 DEBUG Cleaning up

Here's what the gnunet system log says:

May 31 11:49:19-746772 psycstore-sqlite-12929 DEBUG Prepared `DELETE FROM state
WHERE channel_id = (SELECT id FROM channels WHERE pub_key = ?)
      AND (value_current IS NULL OR length(value_current) = 0)
      AND (value_signed IS NULL OR length(value_signed) = 0);' / 0xb8bebbe0: 0
May 31 11:49:19-747149 psycstore-sqlite-12929 DEBUG Prepared `UPDATE state
SET value_signed = value_current
WHERE channel_id = (SELECT id FROM channels WHERE pub_key = ?);' / 0xb8bf5988: 0
May 31 11:49:19-747415 psycstore-sqlite-12929 DEBUG Prepared `DELETE FROM state
WHERE channel_id = (SELECT id FROM channels WHERE pub_key = ?);' / 0xb8bf5a78: 0
May 31 11:49:19-747620 psycstore-sqlite-12929 DEBUG Prepared `INSERT INTO state_sync (channel_id, name, value)
VALUES ((SELECT id FROM channels WHERE pub_key = ?), ?, ?);' / 0xb8bf5b68: 0
May 31 11:49:19-747903 psycstore-sqlite-12929 DEBUG Prepared `INSERT INTO state
 (channel_id, name, value_current, value_signed)
SELECT channel_id, name, value, value
FROM state_sync
WHERE channel_id = (SELECT id FROM channels WHERE pub_key = ?);' / 0xb8bf4fe8: 0
May 31 11:49:19-748146 psycstore-sqlite-12929 DEBUG Prepared `DELETE FROM state_sync
WHERE channel_id = (SELECT id FROM channels WHERE pub_key = ?);' / 0xb8bf52f0: 0
May 31 11:49:19-748409 psycstore-sqlite-12929 DEBUG Prepared `SELECT value_current
FROM state
WHERE channel_id = (SELECT id FROM channels WHERE pub_key = ?)
      AND name = ?;' / 0xb8bf9cd0: 0
May 31 11:49:19-749815 psycstore-sqlite-12929 DEBUG Prepared `SELECT name, value_current
FROM state
WHERE channel_id = (SELECT id FROM channels WHERE pub_key = ?)
      AND (name = ? OR substr(name, 1, ?) = ? || '_');' / 0xb8bf8968: 0
May 31 11:49:19-750213 psycstore-sqlite-12929 DEBUG Prepared `SELECT name, value_signed
FROM state
WHERE channel_id = (SELECT id FROM channels WHERE pub_key = ?) AND value_signed IS NOT NULL;' / 0xb8bfafc8: 0
May 31 11:49:19-750327 psycstore-sqlite-12929 INFO SQLite database running
May 31 11:49:39-720146 core-api-12931 DEBUG Processing message of type 67 and size 40 from core service
May 31 11:49:39-720209 core-api-12931 DEBUG Received notification about connection from `QT80'.

tg

2016-05-31 19:29

manager   ~0010821

Last edited: 2016-05-31 19:32

View 2 revisions

check the IPC logs, it should look something like this for --host-enter
where does it get stuck?

client -> core CORE_INIT (8)
core -> client CORE_INIT_REPLY (40)

client -> social SOCIAL_APP_CONNECT (8)

social -> client SOCIAL_APP_EGO (45)
client WARNING: Ego: Y2MJ2S6X4YJY1RNXYYSQYZ0GD38H2A50JSM5KVJREQZMEJ86KKSG Host One
social -> client SOCIAL_APP_EGO_END (4)

social -> client SOCIAL_APP_PLACE (70)
client WARNING: Host: 7C9DFT92S46EDMPGQ2BQ3XMHX02QRHK3HJNSQZFB57F2V8YAN2Z0
social -> client SOCIAL_APP_PLACE (70)
client WARNING: Guest: 35HMQHF1FJHBWNA9NPTTRW5GNZWJG0WM6QDDS4B1N0G5PDZ526P0
social -> client SOCIAL_APP_PLACE_END (4)

client -> social SOCIAL_HOST_ENTER (108)
social -> psyc PSYC_MASTER_START (40)
psyc -> psycstore PSYCSTORE_COUNTERS_GET (48)
psycstore -> psyc PSYCSTORE_RESULT_COUNTERS (48)
psyc -> social PSYC_MASTER_START_ACK (16)

social -> client SOCIAL_HOST_ENTER_ACK (48)
client WARNING: Host entered: RVZKZ9E17C5M1A8J880FXW6W77V2T607K0473YWT1W2C240TEV70, max_message_id: 0

psyc -> multicast MULTICAST_ORIGIN_START (48)

lynX

2016-05-31 20:27

manager   ~0010822

Your set-up requires me to open far too many windows ;)
Here's the set-up I am using:

user: gnunet-logread -f
system: gnunet-service-arm |& gnunet-logread -x util- -n main
user: gnunet-service-arm |& gnunet-logread -x util- -n main
user: psycit

Output as follows:

] gnunet-logread -f
May 31 19:39:30-616562 main -> core CORE_INIT (8)
May 31 19:39:31-311736 core -> main CORE_INIT_REPLY (40)
May 31 19:39:31-325635 main -> cadet CADET_LOCAL_CONNECT (8)
May 31 19:39:32-583039 core -> main CORE_NOTIFY_CONNECT (40)

system arm:
<shows various things about core, transport, cadet etc>

user arm:

May 31 20:14:50-987834 gns-16358 DEBUG Received 0 records for label `(null)' via namestore monitor
May 31 20:15:31-267992 gns-16358 DEBUG Received 0 records for label `(null)' via namestore monitor
May 31 20:15:32-185269 identity-api-16358 DEBUG Connecting to identity service.
May 31 20:16:11-515757 gns-16358 DEBUG Received 0 records for label `(null)' via namestore monitor
May 31 20:16:12-397646 identity-api-16358 DEBUG Scheduling task to reconnect to identity service in 15 m.
May 31 20:16:51-765743 gns-16358 DEBUG Received 0 records for label `(null)' via namestore monitor
May 31 20:17:32-046055 gns-16358 DEBUG Received 0 records for label `(null)' via namestore monitor

psycit:
Not all of GNS, social and identity services are listening at ./psycit line 63.
(= gnunet-arm is unable to talk to the gnunet-service-arm running on the same uid)

gnunet-identity -d hangs, too. So apparently my way of starting the arm
service manually is incorrect. But how?

Issue History

Date Modified Username Field Change
2016-05-30 07:49 lynX New Issue
2016-05-30 07:49 lynX Status new => assigned
2016-05-30 07:49 lynX Assigned To => tg
2016-05-30 07:49 lynX File Added: psycit
2016-05-30 15:40 tg Note Added: 0010779
2016-05-30 16:50 lynX Note Added: 0010780
2016-05-30 17:10 tg Note Added: 0010781
2016-05-30 18:25 tg Note Edited: 0010781 View Revisions
2016-05-30 18:27 lynX Note Added: 0010782
2016-05-30 18:34 tg Note Added: 0010783
2016-05-30 18:43 lynX Note Added: 0010784
2016-05-30 18:52 tg Note Added: 0010785
2016-05-30 19:07 tg Note Edited: 0010785 View Revisions
2016-05-30 19:07 tg Note Edited: 0010785 View Revisions
2016-05-30 19:10 tg Note Edited: 0010785 View Revisions
2016-05-30 19:37 tg Note Added: 0010788
2016-05-30 19:40 tg Note Added: 0010789
2016-05-30 19:41 tg Note Edited: 0010788 View Revisions
2016-05-30 19:45 tg Note Edited: 0010785 View Revisions
2016-05-30 19:46 tg Note Added: 0010790
2016-05-30 19:46 tg Note Edited: 0010790 View Revisions
2016-05-30 19:49 tg Note Edited: 0010789 View Revisions
2016-05-30 21:05 lynX Note Added: 0010791
2016-05-30 21:26 tg Note Edited: 0010788 View Revisions
2016-05-31 13:53 lynX Note Added: 0010810
2016-05-31 19:29 tg Note Added: 0010821
2016-05-31 19:32 tg Note Edited: 0010821 View Revisions
2016-05-31 20:27 lynX Note Added: 0010822
2016-06-09 14:21 lynX Summary How is the pubsub (gnunet-social) API/CLI supposed to work? => pubsub (gnunet-social) does not work with separated system/user configs
2016-06-09 14:21 lynX Description Updated View Revisions
2018-06-07 01:14 Christian Grothoff Assigned To tg =>
2018-06-07 01:14 Christian Grothoff Status assigned => confirmed
2018-06-07 01:14 Christian Grothoff Target Version SVN HEAD =>
2019-02-12 09:12 Christian Grothoff Project GNUnet => secushare
2019-02-12 09:12 Christian Grothoff Category social service => General
2019-02-12 09:13 Christian Grothoff Category General => social