#1114 get_uid_from_pid() perfoms an improper read
Closed: Fixed None Opened 8 years ago by sgallagh.

    char buf[BUFSIZE];
...
    while ((ret = read(fd, buf, BUFSIZE)) != 0) {
        if (ret == -1) {
            error = errno;
            if (error == EINTR || error == EAGAIN) {
                continue;
            }
            DEBUG(1, ("read failed [%d][%s].\n", error, strerror(error)));
            goto fail_fd;
        }
    }

It's theoretically possible for {{{read()}}} to overrun BUFSIZE here (though it's incredibly unlikely, since we're reading from /proc, so the only way to exploit this file would be a kernel exploit)

This should be rewritten to shrink the request BUFSIZE on subsequent passes to read(). We also need to add an explicit NULL-terminator.


Also, note that buf isn't being advanced after a partial read, so it could be overwriting buf.

We should create a common sss_read() function somewhere that handles reading in a safe way, always.

Fields changed

blockedby: =>
blocking: =>
milestone: NEEDS_TRIAGE => SSSD 1.9.0 NEEDS_TRIAGE

"Nice to have" for 1.9.

milestone: SSSD 1.9.0 NEEDS_TRIAGE => SSSD 1.9.0

Fields changed

rhbz: => 0

Fields changed

feature_milestone: =>
keywords: Coverity => Coverity easyfix

Fixed by 9d7d445

milestone: SSSD 1.9.0 => SSSD 1.9.0 beta 1
resolution: => fixed
status: new => closed

Metadata Update from @sgallagh:
- Issue assigned to jhrozek
- Issue set to the milestone: SSSD 1.9.0 beta 1

3 years ago

SSSD is moving from Pagure to Github. This means that new issues and pull requests
will be accepted only in SSSD's github repository.

This issue has been cloned to Github and is available here:
- https://github.com/SSSD/sssd/issues/2156

If you want to receive further updates on the issue, please navigate to the github issue
and click on subscribe button.

Thank you for understanding. We apologize for all inconvenience.

Login to comment on this ticket.

Metadata