Dear list, for me the agent (v2.2.5) stops responding after one usage or two. It doesn’t seem to recover from this state. Here is what I know:
When resuming after a connection, the agent appears to stop
listening on its socket:
12922 stat("/home/REDACTED/.config/gnupg", {st_mode=S_IFDIR|0700,
st_size=4096, ...}) = 0
12922 pselect6(8, [3 4 6 7], NULL, NULL, {tv_sec=4, tv_nsec=150980}, {[],
8}) = 0 (Timeout)
…
12922 clone(child_stack=0x7fb24f6d7ff0,
flags=CLONE_VM|CLONE_FS|CLONE_FILES|CLONE_SIGHAND|CLONE_THREAD|CLONE_SYSVSEM|CLONE_SETTLS|CLONE_PARENT_SETTID|CLONE_CHILD_CLEARTID,
parent_tidptr=0x7fb24f6d89d0, tls=0x7fb24f6d8700, child_tidptr=0x7fb24f6d89d0)
= 13195
…
13195 exit(0) = ?
13195 +++ exited with 0 +++
12922 <... pselect6 resumed> ) = 0 (Timeout)
12922 wait4(13055, NULL, WNOHANG, NULL) = 0
12922 pselect6(8, [6 7], NULL, NULL, {tv_sec=4, tv_nsec=236006}, {[], 8}) =
0 (Timeout)
12922 wait4(13055, NULL, WNOHANG, NULL) = 0
Where fds 3 and 4 are the gpg and ssh sockets, respectively. 6
and 7 are inotify handles. Agent log:
2018-03-05 11:26:01 gpg-agent[12922] DBG: chan_10 -> BYE
…
2018-03-05 11:26:01 gpg-agent[12922] ssh request handler for sign_request
(13) ready
2018-03-05 11:26:02 gpg-agent[12922] ssh handler 0x7fb24f6d8700 for fd 10
started
2018-03-05 11:26:02 gpg-agent[12922] ssh handler 0x7fb24f6d8700 for fd 10
terminated
2018-03-05 11:26:37 gpg-agent[12922] can't connect my own socket: IPC
connect call failed
2018-03-05 11:26:37 gpg-agent[12922] this process is useless - shutting down
At which point the agent continues indefinitely in the above
loop. It doesn’t recover even when HUP’ed.
gpg-agent runs socket-activated in “supervised” mode. Passing
--disable-check-own-socket makes the problem disappear.
What is going on here?
Best,
Philipp
signature.asc
Description: PGP signature
_______________________________________________ Gnupg-users mailing list [email protected] http://lists.gnupg.org/mailman/listinfo/gnupg-users
