diff options
Diffstat (limited to 'source3/smbd/utmp.c')
| -rw-r--r-- | source3/smbd/utmp.c | 16 |
1 files changed, 6 insertions, 10 deletions
diff --git a/source3/smbd/utmp.c b/source3/smbd/utmp.c index bb48d36946..0dc1917bfe 100644 --- a/source3/smbd/utmp.c +++ b/source3/smbd/utmp.c @@ -472,22 +472,23 @@ static int ut_id_encode(int i, char *fourbyte) int nbase; const char *ut_id_encstr = "0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ"; - fourbyte[0] = 'S'; - fourbyte[1] = 'M'; - /* - * Encode remaining 2 bytes from 'i'. * 'ut_id_encstr' is the character set on which modulo arithmetic is done. * Example: digits would produce the base-10 numbers from '001'. */ nbase = strlen(ut_id_encstr); + fourbyte[0] = ut_id_encstr[i % nbase]; + i /= nbase; + fourbyte[1] = ut_id_encstr[i % nbase]; + i /= nbase; fourbyte[3] = ut_id_encstr[i % nbase]; i /= nbase; fourbyte[2] = ut_id_encstr[i % nbase]; i /= nbase; - return(i); /* 0: good; else overflow */ + /* we do not care about overflows as i is a random number */ + return 0; } #endif /* defined(HAVE_UT_UT_ID) */ @@ -517,11 +518,6 @@ static bool sys_utmp_fill(struct utmp *u, * ut_line: * If size limit proves troublesome, then perhaps use "ut_id_encode()". */ - if (strlen(id_str) > sizeof(u->ut_line)) { - DEBUG(1,("id_str [%s] is too long for %lu char utmp field\n", - id_str, (unsigned long)sizeof(u->ut_line))); - return False; - } utmp_strcpy(u->ut_line, id_str, sizeof(u->ut_line)); #if defined(HAVE_UT_UT_PID) |