On Apr 26, 2011, at 8:56 AM, Cyril Hrubis <[email protected]> wrote:
> Hi!
>> 1. misusage of getpwnam causes ltpuser1 and ltpuser2 to point
>> to same passwd structure, from getpwnam(3):
>> "The return value may point to a static area, and may be overwritten by
>> subsequent calls to getpwent(3)."
>> Fix this by copying the results from returned pointer.
>
> I see, that was caused by replacing the my_getpwnam() by
> SAFE_GETPWNAM().
>
> Anyway my_getpwnam() is wrong hackend interface, we should get rid of
> it. For next time Garrett note that getpwnam() is one of these
> non-reentrant interfaces and do it correctly ;).
Is reentrancy the problem, or the fact that it uses a global variable for the
struct passwd array in the backend?
>
>> +void wait_for_flag(int value)
>> +{
>> + while (1) {
>> + if (*flag == value)
>> + break;
>> + else
>> + sleep(1);
>> + }
>> +}
>> +
>> /*
>> * do_master_child()
>> */
>> @@ -141,10 +158,16 @@ void do_master_child(char **av)
>> char user1name[] = "nobody";
>> char user2name[] = "bin";
>>
>> - struct passwd *ltpuser1, *ltpuser2;
>> + struct passwd *user;
>> + unsigned int uid1, uid2, gid1, gid2;
>
> Technically these should be uid_t and gid_t.
>
>> - ltpuser1 = SAFE_GETPWNAM(NULL, user1name);
>> - ltpuser2 = SAFE_GETPWNAM(NULL, user2name);
>> + user = SAFE_GETPWNAM(NULL, user1name);
>> + uid1 = user->pw_uid;
>> + gid1 = user->pw_gid;
>> +
>> + user = SAFE_GETPWNAM(NULL, user2name);
>> + uid2 = user->pw_uid;
>> + gid2 = user->pw_gid;
>>
>> TEST_EXP_ENOS(exp_enos);
>>
>> @@ -158,11 +181,11 @@ void do_master_child(char **av)
>> tst_brkm(TBROK|TERRNO, cleanup, "Fork failed");
>>
>> if (pid1 == 0) {
>> -
>> - if (setreuid(ltpuser1->pw_uid, ltpuser1->pw_uid) == -1) {
>> + if (setreuid(uid1, gid1) == -1) {
>> perror("setreuid failed (in child)");
>> exit(1);
>> }
>> + *flag = 1;
>> #ifdef UCLINUX
>> if (self_exec(av[0], "") < 0) {
>> perror("self_exec failed");
>> @@ -172,15 +195,18 @@ void do_master_child(char **av)
>> do_child();
>> #endif
>> }
>> - if (setreuid(ltpuser2->pw_uid, ltpuser2->pw_uid) == -1) {
>> + if (setreuid(uid2, gid2) == -1) {
>> perror("seteuid failed");
>> exit(1);
>> }
>>
>> + /* wait until child sets its euid */
>> + wait_for_flag(1);
>> +
>> TEST(kill(pid1, TEST_SIG));
>>
>> /* signal the child that we're done */
>> - *flag = 1;
>> + *flag = 2;
>>
>> if (waitpid(pid1, &status, 0) == -1) {
>> perror("waitpid failed");
>> @@ -209,12 +235,8 @@ void do_child()
>> pid_t my_pid;
>>
>> my_pid = getpid();
>> - while (1) {
>> - if (*flag == 1)
>> - exit(0);
>> - else
>> - sleep(1);
>> - }
>> + wait_for_flag(2);
>> + exit(0);
>> }
>>
>> void setup(void)
>
> And the patch doesn't apply (as your source is older than current git).
>
> Please use latest git sources when creating patches.
Well... The mta was munging with the patches according to Jan. Push comes to
shove, there is pastebin, etc as well that can be used. It's not ideal, but
would allow you to fix your mailer issues and get the patch out in the meantime.
You should really fix that though and/or invest in a free webmail provider like
gmail, yahoo, etc.
Thanks!
-Garrett
------------------------------------------------------------------------------
WhatsUp Gold - Download Free Network Management Software
The most intuitive, comprehensive, and cost-effective network
management toolset available today. Delivers lowest initial
acquisition cost and overall TCO of any competing solution.
http://p.sf.net/sfu/whatsupgold-sd
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list
