[Bug 257368] ggate(d) misbehaving

From: <bugzilla-noreply_at_freebsd.org>
Date: Fri, 23 Jul 2021 23:29:46 +0000
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=257368

            Bug ID: 257368
           Summary: ggate(d) misbehaving
           Product: Base System
           Version: 12.2-STABLE
          Hardware: amd64
                OS: Any
            Status: New
          Severity: Affects Only Me
          Priority: ---
         Component: bin
          Assignee: bugs_at_FreeBSD.org
          Reporter: johannes_at_jo-t.de

Looks like there's a 32 bit integer overflow or underflow glitch in ggated or
ggatec (not sure yet which).

Relevant lines from the log.
ggated first:

[...snipped...]
debug: Received hdr packet.
debug: recv_thread: offset=53686845440 length=114688
debug: disk_thread: offset=53686845440 length=114688
debug: send_thread: offset=53686845440 length=114688
debug: Sent hdr packet.
debug: Sent 114688 bytes (offset=53686845440, size=114688).
debug: Received hdr packet.
debug: recv_thread: offset=16384 length=2147467264
debug: disk_thread: offset=16384 length=2147467264
debug: send_thread: offset=16384 length=2147467264
debug: Sent hdr packet.
debug: Process 2048 exiting.

Notice the suspicious length paramter:
2147467264 = 0x7FFFC000
    114688 = 0x   1C000


Corresponding log from ggatec:

[...snipped...]
debug: Received 114688 bytes (offset=53686845440, size=114688).
debug: Sent hdr packet.
debug: Received hdr packet.
debug: Received data packet.
error: Lost connection 4.
debug: recv_thread: Died.
debug: send_thread: Died.



And ggated promptly allocates an additional 2GB of memory. Line from top:

 PID USERNAME THR PRI NICE   SIZE    RES STATE   C TIME  WCPU COMMAND
2048 root     1   20     0  2077M  1217M spa_na  1 0:05  0.00% ggated

-- 
You are receiving this mail because:
You are the assignee for the bug.
Received on Fri Jul 23 2021 - 23:29:46 UTC

Original text of this message