Opened 11 years ago

Closed 9 years ago

Last modified 7 years ago

#881 closed defect (fixed)

stalled too much writing to 127.0.0.1

Reported by: phobos Owned by:
Priority: Low Milestone: 0.2.1.x-final
Component: Core Tor/Tor Version: 0.2.1.7-alpha
Severity: Keywords:
Cc: phobos, nickm, coderman, arma, karsten, NLonion Actual Points:
Parent ID: Points:
Reviewer: Sponsor:

Description (last modified by phobos)

I've seen this message twice, once while nothing was using Tor, another time while browsing the web via Tor.

[Notice] We stalled too much while trying to write 498 bytes to address "127.0.0.1". If this happens a lot, either som
ething is wrong with your network connection, or something is wrong with theirs. (fd 21, type Socks, state 11, marked at connection_edge.c:
99).

[Automatically added by flyspray2trac: Operating System: Other Linux]

Child Tickets

Change History (13)

comment:1 Changed 11 years ago by phobos

Ran in info loglevel for a while:

Dec 03 22:53:11.907 [Info] connection_edge_reached_eof(): conn (fd 17) reached eof. Closing.
Dec 03 22:53:11.919 [Info] conn_close_if_marked(): Conn (addr "127.0.0.1", fd 17, type Socks, state 11) marked, but wants to flush 498 bytes. (Marked at connection_edge.c:99)
Dec 03 22:53:11.930 [Notice] We stalled too much while trying to write 498 bytes to address "127.0.0.1". If this happens a lot, either something is wrong with your network connection, or something is wrong with theirs. (fd 17, type Socks, state 11, marked at connection_edge.c:99).
Dec 03 22:53:11.942 [Info] connection_edge_process_relay_cell(): data cell dropped, unknown stream.

that last line repeats 31 times and then:

Dec 03 22:53:13.314 [Info] connection_edge_process_relay_cell_not_open(): 'connected' received after 7 seconds.
Dec 03 22:53:13.326 [Info] exit circ (length 3): t0r(open) $9B02AA745B22B3FAB37C84B5E695623DD107A74D(open) $CA57AACA54E93DA1DFE37C7BAC810321AFA12661(open)
Dec 03 22:53:18.658 [Info] connection_edge_process_relay_cell(): end cell (closed normally) dropped, unknown stream.
Dec 03 22:53:19.477 [Info] connection_edge_process_relay_cell(): 20: end cell (closed normally) for stream 4116. Removing stream.
Dec 03 22:53:19.488 [Info] stream_end_reason_to_string(): Reason for ending (518) not recognized.

comment:2 Changed 11 years ago by phobos

And another one today:

Dec 04 14:40:47.325 [Notice] We stalled too much while trying to write 950 bytes to address "127.0.0.1". If this happens a lot, either som
ething is wrong with your network connection, or something is wrong with theirs. (fd 24, type Socks, state 11, marked at connection_edge.c:
99).

Once again, nothing was using Tor other than Vidalia and a hidden service existed.

comment:3 Changed 11 years ago by arma

So no firefox was active at the time?

New theory: your hidden service is making a tunnelled begin_dir connection
to publish its service descriptor. But the circuit fails, so it closes the
stream, and the linked stream has some stuff on its write buffer so it
complains.

comment:4 Changed 11 years ago by hnaparst

I am running a server on Windows XP, and I get this message once every few minutes.
fd is various, type Directory, state 1, marked at main.c: 702.
It happens with both 0.2.1.alpha and 0.2.0.34

comment:5 Changed 10 years ago by NLonion

Me Too... (very often)
Not using the client only running a relay on Windows XP (Tor 0.2.1.19)

aug 29 23:40:06.453 [Notice] We stalled too much while trying to write 107 bytes to address "*.*.*.*". If this happens a lot, either something is wrong with your network connection, or something is wrong with theirs. (fd 656, type Directory, state 1, marked at main.c:702).
aug 29 23:41:32.468 [Notice] We stalled too much while trying to write 122 bytes to address "*.*.*.*". If this happens a lot, either something is wrong with your network connection, or something is wrong with theirs. (fd 10228, type Directory, state 1, marked at main.c:702).
aug 29 23:45:07.656 [Notice] We stalled too much while trying to write 103 bytes to address "*.*.*.*". If this happens a lot, either something is wrong with your network connection, or something is wrong with theirs. (fd 4192, type Directory, state 1, marked at main.c:702).

comment:6 Changed 10 years ago by Sebastian

Are you running a hidden service?

comment:7 Changed 10 years ago by NLonion

No, just a relay (NLonion)
I have set SafeLogging 0 and logged to a file is that is any help to you.

comment:8 Changed 10 years ago by phobos

Is this still occurring with the latest 0.2.1.20 release?

comment:9 Changed 10 years ago by NLonion

Been away for a while. Just installed the latest versions (Vidalia 0.2.5, Tor 0.2.1.20, Qt 4.5.2).
Now we just have to wait.....
CU.

comment:10 Changed 10 years ago by NLonion

Yess, still the same.
I shutdown utorrent just be sure the network was "quiet" but I still get the error message, about 380 in 48 hours.
I've got a log file if you are interrested (only notice and above).
Or I can increase logging if you want me to.
CU.

comment:11 Changed 9 years ago by phobos

Description: modified (diff)

I haven't seen this error in any recent version of tor, no matter how busy tor itself is at the time. Perhaps it's magically fixed.

comment:12 Changed 9 years ago by phobos

Resolution: Nonefixed
Status: newclosed

user disappeared, problem gone away in testing.

comment:13 Changed 7 years ago by nickm

Component: Tor ClientTor
Note: See TracTickets for help on using tickets.