I've got one server, three clients. All gnu-linux, X11. The server is 1.3.4, protocol version 1.3.

After approximately 7-10 days of the server running, the mouse and keyboard events sent to clients become jittery and repetitive. i.e.: You can not type in a password over synergy because the synergy server sends the same keystrokes multiple times. For what its worth they appear to always come in the correct order.

All machines are on the same Gigabit Ethernet network, and no wifi is involved. Network utilization is negligible, and no network problems are occurring. All clients exhibit the same jitter. Restarting clients has no effect. Restarting the synergys process on the server "fixes" the jitter, for another 7 days.

Normally for me, the cpu utilization of synergys on the server spikes to 10% when the cursor moves from server to any client. This doesn't bother me, or seem all that bad. But 7 days later, when the jitter starts, that cpu utilization is nearly 99% on the server.

Memory utilization starts out at 200M, and goes up when large images are copied to the clipboard, but I do not believe this causes the jitter, because I sent several large images via the clipboard push, and the jitter did not recur.

With wwalker's help in IRC, I've been able to identify the above, and I took a coredump of synergys while controlling a client. Here's what it backtraced to: (gdb) bt

0 do_sigwait (set=<value optimized="" out="">, sig=0x7fcaaf886e4c) at ../sysdeps/unix/sysv/linux/sigwait.c:65

1 __sigwait (set=<value optimized="" out="">, sig=0x7fcaaf886e4c) at ../sysdeps/unix/sysv/linux/sigwait.c:100

2 0x000000000045bf2e in ?? ()

3 0x00000030f1006ccb in start_thread (arg=0x7fcaaf887700) at pthread_create.c:301

4 0x00000030f08e0c2d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:115

(gdb) bt full

0 do_sigwait (set=<value optimized="" out="">, sig=0x7fcaaf886e4c) at ../sysdeps/unix/sysv/linux/sigwait.c:65

    ret = <value optimized out>
    tmpset = {__val = {0 <repeats 16 times>}}

1 __sigwait (set=<value optimized="" out="">, sig=0x7fcaaf886e4c) at ../sysdeps/unix/sysv/linux/sigwait.c:100

    oldtype = 0
    result = -4

2 0x000000000045bf2e in ?? ()

No symbol table info available.

3 0x00000030f1006ccb in start_thread (arg=0x7fcaaf887700) at pthread_create.c:301

    pd = 0x7fcaaf887700
    now = <value optimized out>
    unwind_buf = {cancel_jmp_buf = {{jmp_buf = {140508505077504, 8067460418120327409, 140736027569328, 140508505078208, 0, 3, -8043359042187943695, 
            8040194725672919281}, mask_was_saved = 0}}, priv = {pad = {0x0, 0x0, 0x0, 0x0}, data = {prev = 0x0, cleanup = 0x0, canceltype = 0}}}
    not_first_call = <value optimized out>
    pagesize_m1 = <value optimized out>
    sp = <value optimized out>
    freesize = <value optimized out>

4 0x00000030f08e0c2d in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:115

No locals. (gdb)

What do I need to do to make synergy more stable? Is there any better debugging info I can provide to help identify a bug, and get it fixed?

I'd like to get it to the point where the synergys process can start at boot time, and stay running for several months at a time, until I have to reboot for hardware maintenance or kernel upgrade.

Thanks, for the absolutely outstanding software! I appreciate it, sincerely.