Home / server / Questions / 211548
Accepted
Istvan
Asked: 2010-12-12 05:16:34 +0800 CST

Ctrl + c is not working for certain applications on Linux

  • 772

It is a really weird problem, but on new systems(Fedora, Ubuntu) ctrl+c has no effect for certain tools:

if I execute yum list which runs for almost a minute I can't interrupt to run with ctrl+c

$time yum list >/dev/null
^C^C^C^C^C^C^C^C^C

The command execution won't stop.

However to interrupt find is possible.

$ time find / >/dev/null 2>&1
^C

real    0m0.741s
user    0m0.033s
sys     0m0.124s

I am curious what is causing this.

I have to following settings:

$ stty -a
speed 38400 baud; rows 36; columns 158; line = 0;
intr = ^C; quit = ^\; erase = ^?; kill = ^U; eof = ^D; eol = <undef>; eol2 = <undef>; swtch = <undef>; start = ^Q; stop = ^S; susp = ^Z; rprnt = ^R;
werase = ^W; lnext = ^V; flush = ^O; min = 1; time = 0;
-parenb -parodd cs8 -hupcl -cstopb cread -clocal -crtscts
-ignbrk -brkint -ignpar -parmrk -inpck -istrip -inlcr -igncr icrnl ixon -ixoff -iuclc -ixany -imaxbel -iutf8
opost -olcuc -ocrnl onlcr -onocr -onlret -ofill -ofdel nl0 cr0 tab0 bs0 vt0 ff0
isig icanon iexten echo echoe echok -echonl -noflsh -xcase -tostop -echoprt echoctl echoke

I couldn't find the description of the problem yet, I appreciate if somebody would be that kind to shed some light.

Thanks in advance.

linux command-line-interface ssh terminal

6 Answers

  1. Some applications trap SIGINT because of strange interactions with other libraries. You can try sending them a SIGQUIT instead (via Ctrl\ as given by your stty output).

    • 9

  2. Some applications just trap SIGINT as Ignacio mentioned, others capture all keyboard input.

    If C-c doesn't work, you may try the already mentioned C-\ and if this doesn't work, then just try to background the process: C-z. and then kill it with kill -s SIGKILL <pid>

    • 2
  3. Best Answer

    This is a yum bug. It seems there are more and more linux developers think it is a good idea to ignore standard signal handling and make their app to not react to standard signals.

    man 7 signal

       Standard Signals
       Linux  supports the standard signals listed below.  Several signal numbers are architecture-dependent, as indicated in the "Value" col-
       umn.  (Where three values are given, the first one is usually valid for alpha and sparc, the middle one for ix86, ia64, ppc, s390,  arm
       and sh, and the last one for mips.  A - denotes that a signal is absent on the corresponding architecture.)

       First the signals described in the original POSIX.1-1990 standard.

       Signal     Value     Action   Comment
       ----------------------------------------------------------------------
       SIGHUP        1       Term    Hangup detected on controlling terminal
                                     or death of controlling process
       SIGINT        2       Term    Interrupt from keyboard
       SIGQUIT       3       Core    Quit from keyboard
       SIGILL        4       Core    Illegal Instruction
       SIGABRT       6       Core    Abort signal from abort(3)
       SIGFPE        8       Core    Floating point exception
       SIGKILL       9       Term    Kill signal
       SIGSEGV      11       Core    Invalid memory reference
       SIGPIPE      13       Term    Broken pipe: write to pipe with no
                                     readers
       SIGALRM      14       Term    Timer signal from alarm(2)
       SIGTERM      15       Term    Termination signal
       SIGUSR1   30,10,16    Term    User-defined signal 1
       SIGUSR2   31,12,17    Term    User-defined signal 2
       SIGCHLD   20,17,18    Ign     Child stopped or terminated
       SIGCONT   19,18,25    Cont    Continue if stopped
       SIGSTOP   17,19,23    Stop    Stop process
       SIGTSTP   18,20,24    Stop    Stop typed at tty
       SIGTTIN   21,21,26    Stop    tty input for background process
       SIGTTOU   22,22,27    Stop    tty output for background process

    This is a bad habit.

    • 1

  4. Definitely related to signal handling in the command you're calling. See a number of open yum bugs around the handling:

    Red Hat Bugzilla – Bug List

    https://bugzilla.redhat.com/buglist.cgi?quicksearch=yum+Ctrl-C

    Seems like Ctrl-C (SIGINT) might have been used to control other behavior (skipping to the next mirror) rather than the usual intention (killing the process).

    Re: why SIGQUIT doesn't seem to do anything useful -- there may not be a defined handler.

    • 1

  5. This usually happens when the application is in a state where it's not actually able to respond to signals because it's not able to resume running from the CPU scheduler queue. A good example is when the application is hung waiting for a blocking operation (synchronous disk I/O in the main thread, swap in/out, blocking network I/O, etc.) Given that this is yum, I'm guessing that it's timing out connecting to one or more of the sources defined in your repository/mirrorlist configs, but it could easily be a disk I/O issue accessing its caches, the RPM database (including BDB corruption), or any number of other things.

    A good way to test behavior this with any arbitrary application: hard-mount an NFS share, then pull the NFS server offline, then try to use any program that tries to open that directory (find or /bin/ls are good examples). It won't respond to anything beyond a SIGKILL while it waits for the kernel to figure out what's up with the share.

    • 0

  6. Actually, rpmlib, used by yum, is grabbing and ignoring the ctrl-c signal handler. There's only so much can be done about this at yum's level. It's annoying, but I'm not sure it's worth getting all worked up about. Newer and newer versions of Fedora have successfully improved the handling of this, to the point where yum under Rawhide (which will become F15) does this for your test case:

    $ time yum list >/dev/null
^CTraceback (most recent call last):
  File "/usr/lib64/python2.7/site.py", line 553, in <module>
    main()
  File "/usr/lib64/python2.7/site.py", line 544, in main
    execsitecustomize()
  File "/usr/lib64/python2.7/site.py", line 500, in execsitecustomize
    import sitecustomize
KeyboardInterrupt

real    0m0.164s
user    0m0.081s
sys     0m0.073s

    It still will not let you interrupt changes to the RPM database with SIGINT, which is hard to really argue with.

    • 0