<div dir="ltr">I am running into the exact same CRIU failure with a mysqld server running in a container:<div><br></div><div><div>(00.007793) Error (proc_parse.c:404): Can't handle non-regular mapping on 26317's map 7ff7022cc000</div></div><div><br></div><div>for the following segment:</div><div><br></div><div><div>7ff7022cc000-7ff7022ce000 rw-s 00000000 00:0a 72033 /[aio] (deleted)</div></div><div><br></div><div>I am using CRIU 1.6 on a 3.18.0 kernel. Below is the output of strace. Kernel successfully opens the file:</div><div><br></div><div><div># criu --version</div><div>Version: 1.6</div><div>GitID: v1.6-5-g5e1a8a2</div><div># criu check</div><div>Error (cr-check.c:665): AIO remap doesn't work properly</div><div>Error (cr-check.c:681): fdinfo doesn't contain the lock field</div><div># zcat /proc/config.gz | grep AIO</div><div>CONFIG_AIO=y</div><div># CONFIG_DELL_WMI_AIO is not set</div></div><div># strace ./aio_test /proc/26317/map_files/7ff7022cc000-7ff7022ce000<br></div><div><div>execve("./aio_test", ["./aio_test", "/proc/26317/map_files/7ff7022cc0"...], [/* 17 vars */]) = 0</div><div>uname({sys="Linux", node="saied-cr-vm", ...}) = 0</div><div>brk(0) = 0x203d000</div><div>brk(0x203e1c0) = 0x203e1c0</div><div>arch_prctl(ARCH_SET_FS, 0x203d880) = 0</div><div>readlink("/proc/self/exe", "/home/saied/aio_test", 4096) = 20</div><div>brk(0x205f1c0) = 0x205f1c0</div><div>brk(0x2060000) = 0x2060000</div><div>access("/etc/ld.so.nohwcap", F_OK) = -1 ENOENT (No such file or directory)</div><div>open("/proc/26317/map_files/7ff7022cc000-7ff7022ce000", O_RDONLY) = 3</div><div>exit_group(3) = ?</div><div>+++ exited with 3 +++</div></div><div><br></div><div>Any ideas?</div><div><br></div><div>--Saied</div><div><br></div></div><div class="gmail_extra"><br><div class="gmail_quote">On Thu, Jun 4, 2015 at 2:18 AM, Pavel Emelyanov <span dir="ltr"><<a href="mailto:xemul@parallels.com" target="_blank">xemul@parallels.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">On 06/03/2015 08:42 PM, Ross Boucher wrote:<br>
> This is a new instance of the same process -- the dump is the same, but the map id that's failing now is:<br>
> 7f81502d9000 (just to explain that the other id won't appear in this log)<br>
<br>
The bogus mapping is<br>
<br>
7f81502d9000-7f81502dc000 rw-s 00000000 00:0b 1635352 /[aio] (deleted)<br>
<br>
OK, this is the AIO ring. Strange, CRIU should support it. I have several<br>
questions.<br>
<br>
1. What's the kernel version you're using?<br>
2. Can you run the "criu check", what would it show?<br>
3. Also, please, take this proggie<br>
<br>
#include <stdio.h><br>
<br>
int main(int argc, char **argv)<br>
{<br>
open(argv[1], 0);<br>
}<br>
<br>
<br>
and strace what it does on the /proc/$nginx_pid/map_files/$aio_mapping. The<br>
correct behavior should be -ENXIO reported by the kernel. By seeing this<br>
CRIU finds out that the mapping in question is AIO. In some older kernels<br>
this rung used to be open-able, but on those kernels AIO rings were not<br>
remap-able, so CRIU didn't support AIO-s on them. Maybe that's your case?<br>
<span class="HOEnZb"><font color="#888888"><br>
-- Pavel<br>
</font></span><div class="HOEnZb"><div class="h5">_______________________________________________<br>
CRIU mailing list<br>
<a href="mailto:CRIU@openvz.org">CRIU@openvz.org</a><br>
<a href="https://lists.openvz.org/mailman/listinfo/criu" target="_blank">https://lists.openvz.org/mailman/listinfo/criu</a><br>
</div></div></blockquote></div><br></div>