shadow/src/groupadd.c

655 lines
17 KiB
C
Raw Normal View History

/*
* Copyright (c) 1991 - 1993, Julianne Frances Haugh
* Copyright (c) 1996 - 2000, Marek Michałkiewicz
* Copyright (c) 2000 - 2006, Tomasz Kłoczko
* Copyright (c) 2007 - 2011, Nicolas François
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. The name of the copyright holders or contributors may not be used to
* endorse or promote products derived from this software without
* specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
* ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
* LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
* PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
* HOLDERS OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
* SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
* LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
* OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include <config.h>
#ident "$Id$"
#include <ctype.h>
#include <fcntl.h>
#include <getopt.h>
#include <grp.h>
#include <stdio.h>
#include <sys/types.h>
#ifdef ACCT_TOOLS_SETUID
#ifdef USE_PAM
#include "pam_defs.h"
#include <pwd.h>
#endif /* USE_PAM */
#endif /* ACCT_TOOLS_SETUID */
#include "chkname.h"
#include "defines.h"
#include "getdef.h"
#include "groupio.h"
#include "nscd.h"
Flush sssd caches in addition to nscd caches Some distributions, notably Fedora, have the following order of nsswitch modules by default: passwd: sss files group: sss files The advantage of serving local users through SSSD is that the nss_sss module has a fast mmapped-cache that speeds up NSS lookups compared to accessing the disk an opening the files on each NSS request. Traditionally, this has been done with the help of nscd, but using nscd in parallel with sssd is cumbersome, as both SSSD and nscd use their own independent caching, so using nscd in setups where sssd is also serving users from some remote domain (LDAP, AD, ...) can result in a bit of unpredictability. More details about why Fedora chose to use sss before files can be found on e.g.: https://fedoraproject.org//wiki/Changes/SSSDCacheForLocalUsers or: https://docs.pagure.org/SSSD.sssd/design_pages/files_provider.html Now, even though sssd watches the passwd and group files with the help of inotify, there can still be a small window where someone requests a user or a group, finds that it doesn't exist, adds the entry and checks again. Without some support in shadow-utils that would explicitly drop the sssd caches, the inotify watch can fire a little late, so a combination of commands like this: getent passwd user || useradd user; getent passwd user can result in the second getent passwd not finding the newly added user as the racy behaviour might still return the cached negative hit from the first getent passwd. This patch more or less copies the already existing support that shadow-utils had for dropping nscd caches, except using the "sss_cache" tool that sssd ships.
2018-09-12 17:52:11 +05:30
#include "sssd.h"
#include "prototypes.h"
#ifdef SHADOWGRP
#include "sgroupio.h"
#endif
/*
* exit status values
*/
/*@-exitarg@*/
#define E_SUCCESS 0 /* success */
#define E_USAGE 2 /* invalid command syntax */
#define E_BAD_ARG 3 /* invalid argument to option */
#define E_GID_IN_USE 4 /* gid not unique (when -o not used) */
#define E_NAME_IN_USE 9 /* group name not unique */
#define E_GRP_UPDATE 10 /* can't update group file */
/*
* Global variables
*/
const char *Prog;
FILE *shadow_logfd = NULL;
static /*@null@*/char *group_name;
static gid_t group_id;
static /*@null@*/char *group_passwd;
static /*@null@*/char *empty_list = NULL;
2016-05-15 19:19:39 +05:30
static const char *prefix = "";
static char *user_list;
2016-05-15 19:19:39 +05:30
static bool oflg = false; /* permit non-unique group ID to be specified with -g */
static bool gflg = false; /* ID value for the new group */
static bool fflg = false; /* if group already exists, do nothing and exit(0) */
static bool rflg = false; /* create a system account */
static bool pflg = false; /* new encrypted password */
#ifdef SHADOWGRP
static bool is_shadow_grp;
#endif
/* local function prototypes */
static /*@noreturn@*/void usage (int status);
static void new_grent (struct group *grent);
#ifdef SHADOWGRP
static void new_sgent (struct sgrp *sgent);
#endif
static void grp_update (void);
static void check_new_name (void);
static void close_files (void);
static void open_files (void);
static void process_flags (int argc, char **argv);
static void check_flags (void);
static void check_perms (void);
/*
* usage - display usage message and exit
*/
static /*@noreturn@*/void usage (int status)
{
FILE *usageout = (E_SUCCESS != status) ? stderr : stdout;
(void) fprintf (usageout,
_("Usage: %s [options] GROUP\n"
"\n"
"Options:\n"),
Prog);
(void) fputs (_(" -f, --force exit successfully if the group already exists,\n"
" and cancel -g if the GID is already used\n"), usageout);
(void) fputs (_(" -g, --gid GID use GID for the new group\n"), usageout);
(void) fputs (_(" -h, --help display this help message and exit\n"), usageout);
(void) fputs (_(" -K, --key KEY=VALUE override /etc/login.defs defaults\n"), usageout);
(void) fputs (_(" -o, --non-unique allow to create groups with duplicate\n"
" (non-unique) GID\n"), usageout);
(void) fputs (_(" -p, --password PASSWORD use this encrypted password for the new group\n"), usageout);
(void) fputs (_(" -r, --system create a system account\n"), usageout);
(void) fputs (_(" -R, --root CHROOT_DIR directory to chroot into\n"), usageout);
(void) fputs (_(" -P, --prefix PREFIX_DI directory prefix\n"), usageout);
(void) fputs (_(" -U, --users USERS list of user members of this group\n"), usageout);
(void) fputs ("\n", usageout);
exit (status);
}
/*
* new_grent - initialize the values in a group file entry
*
* new_grent() takes all of the values that have been entered and fills
* in a (struct group) with them.
*/
static void new_grent (struct group *grent)
{
memzero (grent, sizeof *grent);
grent->gr_name = group_name;
if (pflg) {
grent->gr_passwd = group_passwd;
} else {
grent->gr_passwd = SHADOW_PASSWD_STRING; /* XXX warning: const */
}
grent->gr_gid = group_id;
grent->gr_mem = &empty_list;
}
#ifdef SHADOWGRP
/*
* new_sgent - initialize the values in a shadow group file entry
*
* new_sgent() takes all of the values that have been entered and fills
* in a (struct sgrp) with them.
*/
static void new_sgent (struct sgrp *sgent)
{
memzero (sgent, sizeof *sgent);
sgent->sg_name = group_name;
if (pflg) {
sgent->sg_passwd = group_passwd;
} else {
sgent->sg_passwd = "!"; /* XXX warning: const */
}
sgent->sg_adm = &empty_list;
sgent->sg_mem = &empty_list;
}
#endif /* SHADOWGRP */
/*
* grp_update - add new group file entries
*
* grp_update() writes the new records to the group files.
*/
static void grp_update (void)
{
struct group grp;
#ifdef SHADOWGRP
struct sgrp sgrp;
#endif /* SHADOWGRP */
/*
* To add the group, we need to update /etc/group.
* Make sure failures will be reported.
*/
add_cleanup (cleanup_report_add_group_group, group_name);
#ifdef SHADOWGRP
if (is_shadow_grp) {
/* We also need to update /etc/gshadow */
add_cleanup (cleanup_report_add_group_gshadow, group_name);
}
#endif
/*
* Create the initial entries for this new group.
*/
new_grent (&grp);
#ifdef SHADOWGRP
new_sgent (&sgrp);
if (is_shadow_grp && pflg) {
2008-01-05 22:14:51 +05:30
grp.gr_passwd = SHADOW_PASSWD_STRING; /* XXX warning: const */
}
#endif /* SHADOWGRP */
if (user_list) {
char *token;
token = strtok(user_list, ",");
while (token) {
if (prefix_getpwnam (token) == NULL) {
fprintf (stderr, _("Invalid member username %s\n"), token);
exit (E_GRP_UPDATE);
}
grp.gr_mem = add_list(grp.gr_mem, token);
token = strtok(NULL, ",");
}
}
/*
* Write out the new group file entry.
*/
if (gr_update (&grp) == 0) {
fprintf (stderr,
_("%s: failed to prepare the new %s entry '%s'\n"),
Prog, gr_dbname (), grp.gr_name);
exit (E_GRP_UPDATE);
}
#ifdef SHADOWGRP
/*
* Write out the new shadow group entries as well.
*/
if (is_shadow_grp && (sgr_update (&sgrp) == 0)) {
fprintf (stderr,
_("%s: failed to prepare the new %s entry '%s'\n"),
Prog, sgr_dbname (), sgrp.sg_name);
exit (E_GRP_UPDATE);
}
#endif /* SHADOWGRP */
}
/*
* check_new_name - check the new name for validity
*
* check_new_name() insures that the new name doesn't contain any
* illegal characters.
*/
static void check_new_name (void)
{
if (is_valid_group_name (group_name)) {
return;
2007-12-28 15:42:09 +05:30
}
/*
* All invalid group names land here.
*/
fprintf (stderr, _("%s: '%s' is not a valid group name\n"),
2007-12-29 02:05:05 +05:30
Prog, group_name);
exit (E_BAD_ARG);
}
/*
* close_files - close all of the files that were opened
*
* close_files() closes all of the files that were opened for this new
* group. This causes any modified entries to be written out.
*/
static void close_files (void)
{
/* First, write the changes in the regular group database */
if (gr_close () == 0) {
fprintf (stderr,
_("%s: failure while writing changes to %s\n"),
Prog, gr_dbname ());
exit (E_GRP_UPDATE);
}
#ifdef WITH_AUDIT
audit_logger (AUDIT_ADD_GROUP, Prog,
"adding group to /etc/group",
group_name, (unsigned int) group_id,
SHADOW_AUDIT_SUCCESS);
#endif
SYSLOG ((LOG_INFO, "group added to %s: name=%s, GID=%u",
gr_dbname (), group_name, (unsigned int) group_id));
del_cleanup (cleanup_report_add_group_group);
cleanup_unlock_group (NULL);
del_cleanup (cleanup_unlock_group);
/* Now, write the changes in the shadow database */
#ifdef SHADOWGRP
2007-12-28 15:42:09 +05:30
if (is_shadow_grp) {
if (sgr_close () == 0) {
fprintf (stderr,
_("%s: failure while writing changes to %s\n"),
Prog, sgr_dbname ());
exit (E_GRP_UPDATE);
}
#ifdef WITH_AUDIT
audit_logger (AUDIT_ADD_GROUP, Prog,
"adding group to /etc/gshadow",
group_name, (unsigned int) group_id,
SHADOW_AUDIT_SUCCESS);
#endif
SYSLOG ((LOG_INFO, "group added to %s: name=%s",
sgr_dbname (), group_name));
del_cleanup (cleanup_report_add_group_gshadow);
cleanup_unlock_gshadow (NULL);
del_cleanup (cleanup_unlock_gshadow);
2007-12-28 15:42:09 +05:30
}
#endif /* SHADOWGRP */
/* Report success at the system level */
#ifdef WITH_AUDIT
audit_logger (AUDIT_ADD_GROUP, Prog,
"",
group_name, (unsigned int) group_id,
SHADOW_AUDIT_SUCCESS);
#endif
SYSLOG ((LOG_INFO, "new group: name=%s, GID=%u",
group_name, (unsigned int) group_id));
del_cleanup (cleanup_report_add_group);
}
/*
* open_files - lock and open the group files
*
* open_files() opens the two group files.
*/
static void open_files (void)
{
/* First, lock the databases */
if (gr_lock () == 0) {
fprintf (stderr,
_("%s: cannot lock %s; try again later.\n"),
Prog, gr_dbname ());
exit (E_GRP_UPDATE);
}
add_cleanup (cleanup_unlock_group, NULL);
#ifdef SHADOWGRP
if (is_shadow_grp) {
if (sgr_lock () == 0) {
fprintf (stderr,
_("%s: cannot lock %s; try again later.\n"),
Prog, sgr_dbname ());
exit (E_GRP_UPDATE);
}
add_cleanup (cleanup_unlock_gshadow, NULL);
}
#endif /* SHADOWGRP */
/*
* Now if the group is not added, it's our fault.
* Make sure failures will be reported.
*/
add_cleanup (cleanup_report_add_group, group_name);
2011-07-09 01:21:32 +05:30
/* And now open the databases */
if (gr_open (O_CREAT | O_RDWR) == 0) {
fprintf (stderr, _("%s: cannot open %s\n"), Prog, gr_dbname ());
SYSLOG ((LOG_WARN, "cannot open %s", gr_dbname ()));
exit (E_GRP_UPDATE);
}
#ifdef SHADOWGRP
if (is_shadow_grp) {
if (sgr_open (O_CREAT | O_RDWR) == 0) {
fprintf (stderr,
_("%s: cannot open %s\n"),
Prog, sgr_dbname ());
SYSLOG ((LOG_WARN, "cannot open %s", sgr_dbname ()));
exit (E_GRP_UPDATE);
}
2007-12-28 15:42:09 +05:30
}
#endif /* SHADOWGRP */
}
/*
* process_flags - parse the command line options
*
* It will not return if an error is encountered.
*/
static void process_flags (int argc, char **argv)
{
/*
* Parse the command line options.
*/
char *cp;
int c;
static struct option long_options[] = {
{"force", no_argument, NULL, 'f'},
{"gid", required_argument, NULL, 'g'},
{"help", no_argument, NULL, 'h'},
{"key", required_argument, NULL, 'K'},
{"non-unique", no_argument, NULL, 'o'},
{"password", required_argument, NULL, 'p'},
{"system", no_argument, NULL, 'r'},
{"root", required_argument, NULL, 'R'},
2016-05-15 19:19:39 +05:30
{"prefix", required_argument, NULL, 'P'},
{"users", required_argument, NULL, 'U'},
{NULL, 0, NULL, '\0'}
};
while ((c = getopt_long (argc, argv, "fg:hK:op:rR:P:U:",
long_options, NULL)) != -1) {
switch (c) {
case 'f':
/*
* "force" - do nothing, just exit(0), if the
* specified group already exists. With -g, if
* specified gid already exists, choose another
* (unique) gid (turn off -g). Based on the RedHat's
* patch from shadow-utils-970616-9.
*/
fflg = true;
break;
case 'g':
gflg = true;
if ( (get_gid (optarg, &group_id) == 0)
|| (group_id == (gid_t)-1)) {
fprintf (stderr,
_("%s: invalid group ID '%s'\n"),
Prog, optarg);
exit (E_BAD_ARG);
}
break;
case 'h':
usage (E_SUCCESS);
/*@notreached@*/break;
case 'K':
/*
* override login.defs defaults (-K name=value)
* example: -K GID_MIN=100 -K GID_MAX=499
* note: -K GID_MIN=10,GID_MAX=499 doesn't work yet
*/
cp = strchr (optarg, '=');
if (NULL == cp) {
fprintf (stderr,
* lib/groupio.c, lib/groupio.h, lib/pwio.c, lib/pwio.h, lib/sgroupio.c, lib/sgroupio.h, lib/shadowio.c, lib/shadowio.h: Added *_dbname() functions to retrieve the name of the databases. * lib/groupio.c, lib/groupio.h, lib/pwio.c, lib/pwio.h, lib/sgroupio.c, lib/sgroupio.h, lib/shadowio.c, lib/shadowio.h: *_name() functions renamed *setname(). * src/grpck.c, src/pwck.c: Likewise. * lib/groupio.h, lib/pwio.h, lib/sgroupio.h, lib/shadowio.h: Added the name of the arguments to the prototypes. * src/chage, src/chfn.c, src/chgpasswd.c, src/chpasswd.c, src/chsh.c, src/gpasswd.c, src/groupadd.c, src/groupdel.c, src/groupmod.c, src/grpck.c, src/grpconv.c, src/grpunconv.c, src/newusers.c, src/passwd.c, src/pwck.c, src/pwconv.c, src/pwunconv.c, src/useradd.c, src/userdel.c, src/usermod.c: Harmonize the erro & syslog messages in case of failure of the *_lock(), *_open(), *_close(), *_unlock(), *_remove() functions. * src/chgpasswd.c, src/chpasswd.c, src/usermod.c: Avoid capitalized messages. * src/chpasswd.c, src/useradd.c, src/usermod.c: Harmonize messages in case of inexistent entries. * src/usermod.c: Harmonize messages in case of already existing entries. * src/newusers.c, src/useradd.c: Simplify PAM error handling. * src/useradd.c: Report failures to unlock files (stderr, syslog, and audit). But do not fail (continue). * src/useradd.c (open_files): Do not report to syslog & audit failures to lock or open the databases. This might be harmless, and the logs were not already informed that a change was requested. * src/usermod.c: It's not the account which is unlocked, but its password.
2008-08-06 21:27:31 +05:30
_("%s: -K requires KEY=VALUE\n"),
2007-12-29 02:05:05 +05:30
Prog);
exit (E_BAD_ARG);
}
/* terminate name, point to value */
*cp++ = '\0';
if (putdef_str (optarg, cp) < 0) {
exit (E_BAD_ARG);
}
break;
case 'o':
oflg = true;
break;
case 'p':
pflg = true;
group_passwd = optarg;
break;
case 'r':
rflg = true;
break;
case 'R': /* no-op, handled in process_root_flag () */
break;
2016-05-15 19:19:39 +05:30
case 'P': /* no-op, handled in process_prefix_flag () */
break;
case 'U':
user_list = optarg;
break;
default:
usage (E_USAGE);
}
}
/*
* Check the flags consistency
*/
if (optind != argc - 1) {
usage (E_USAGE);
}
group_name = argv[optind];
check_flags ();
}
/*
* check_flags - check flags and parameters consistency
*
* It will not return if an error is encountered.
*/
static void check_flags (void)
{
/* -o does not make sense without -g */
if (oflg && !gflg) {
usage (E_USAGE);
}
check_new_name ();
/*
* Check if the group already exist.
*/
/* local, no need for xgetgrnam */
2016-05-15 19:19:39 +05:30
if (prefix_getgrnam (group_name) != NULL) {
/* The group already exist */
if (fflg) {
/* OK, no need to do anything */
exit (E_SUCCESS);
}
2008-12-23 03:33:34 +05:30
fprintf (stderr,
_("%s: group '%s' already exists\n"),
Prog, group_name);
exit (E_NAME_IN_USE);
}
2016-05-15 19:19:39 +05:30
if (gflg && (prefix_getgrgid (group_id) != NULL)) {
/* A GID was specified, and a group already exist with that GID
* - either we will use this GID anyway (-o)
* - either we ignore the specified GID and
2008-02-20 02:48:04 +05:30
* we will use another one (-f)
* - either it is a failure
*/
if (oflg) {
/* Continue with this GID */
} else if (fflg) {
/* Turn off -g, we can use any GID */
gflg = false;
} else {
2008-12-23 03:33:34 +05:30
fprintf (stderr,
_("%s: GID '%lu' already exists\n"),
* lib/groupio.c, lib/groupio.h, lib/pwio.c, lib/pwio.h, lib/sgroupio.c, lib/sgroupio.h, lib/shadowio.c, lib/shadowio.h: Added *_dbname() functions to retrieve the name of the databases. * lib/groupio.c, lib/groupio.h, lib/pwio.c, lib/pwio.h, lib/sgroupio.c, lib/sgroupio.h, lib/shadowio.c, lib/shadowio.h: *_name() functions renamed *setname(). * src/grpck.c, src/pwck.c: Likewise. * lib/groupio.h, lib/pwio.h, lib/sgroupio.h, lib/shadowio.h: Added the name of the arguments to the prototypes. * src/chage, src/chfn.c, src/chgpasswd.c, src/chpasswd.c, src/chsh.c, src/gpasswd.c, src/groupadd.c, src/groupdel.c, src/groupmod.c, src/grpck.c, src/grpconv.c, src/grpunconv.c, src/newusers.c, src/passwd.c, src/pwck.c, src/pwconv.c, src/pwunconv.c, src/useradd.c, src/userdel.c, src/usermod.c: Harmonize the erro & syslog messages in case of failure of the *_lock(), *_open(), *_close(), *_unlock(), *_remove() functions. * src/chgpasswd.c, src/chpasswd.c, src/usermod.c: Avoid capitalized messages. * src/chpasswd.c, src/useradd.c, src/usermod.c: Harmonize messages in case of inexistent entries. * src/usermod.c: Harmonize messages in case of already existing entries. * src/newusers.c, src/useradd.c: Simplify PAM error handling. * src/useradd.c: Report failures to unlock files (stderr, syslog, and audit). But do not fail (continue). * src/useradd.c (open_files): Do not report to syslog & audit failures to lock or open the databases. This might be harmless, and the logs were not already informed that a change was requested. * src/usermod.c: It's not the account which is unlocked, but its password.
2008-08-06 21:27:31 +05:30
Prog, (unsigned long int) group_id);
exit (E_GID_IN_USE);
}
}
}
/*
* check_perms - check if the caller is allowed to add a group
*
* With PAM support, the setuid bit can be set on groupadd to allow
* non-root users to groups.
* Without PAM support, only users who can write in the group databases
* can add groups.
*
* It will not return if the user is not allowed.
*/
static void check_perms (void)
{
#ifdef ACCT_TOOLS_SETUID
#ifdef USE_PAM
pam_handle_t *pamh = NULL;
int retval;
struct passwd *pampw;
pampw = getpwuid (getuid ()); /* local, no need for xgetpwuid */
if (NULL == pampw) {
fprintf (stderr,
_("%s: Cannot determine your user name.\n"),
Prog);
exit (1);
}
retval = pam_start ("groupadd", pampw->pw_name, &conv, &pamh);
if (PAM_SUCCESS == retval) {
retval = pam_authenticate (pamh, 0);
}
if (PAM_SUCCESS == retval) {
retval = pam_acct_mgmt (pamh, 0);
}
if (PAM_SUCCESS != retval) {
fprintf (stderr, _("%s: PAM: %s\n"),
Prog, pam_strerror (pamh, retval));
SYSLOG((LOG_ERR, "%s", pam_strerror (pamh, retval)));
if (NULL != pamh) {
(void) pam_end (pamh, retval);
}
exit (1);
}
(void) pam_end (pamh, retval);
#endif /* USE_PAM */
#endif /* ACCT_TOOLS_SETUID */
}
/*
* main - groupadd command
*/
int main (int argc, char **argv)
{
/*
* Get my name so that I can use it to report errors.
*/
Prog = Basename (argv[0]);
shadow_logfd = stderr;
(void) setlocale (LC_ALL, "");
(void) bindtextdomain (PACKAGE, LOCALEDIR);
(void) textdomain (PACKAGE);
process_root_flag ("-R", argc, argv);
2016-05-15 19:19:39 +05:30
prefix = process_prefix_flag ("-P", argc, argv);
OPENLOG ("groupadd");
#ifdef WITH_AUDIT
audit_help_open ();
#endif
if (atexit (do_cleanups) != 0) {
fprintf (stderr,
_("%s: Cannot setup cleanup service.\n"),
Prog);
exit (1);
}
/*
* Parse the command line options.
*/
process_flags (argc, argv);
check_perms ();
#ifdef SHADOWGRP
is_shadow_grp = sgr_file_present ();
#endif
/*
* Do the hard stuff - open the files, create the group entries,
* then close and update the files.
*/
open_files ();
if (!gflg) {
if (find_new_gid (rflg, &group_id, NULL) < 0) {
exit (E_GID_IN_USE);
}
2007-12-28 15:42:09 +05:30
}
grp_update ();
close_files ();
nscd_flush_cache ("group");
Flush sssd caches in addition to nscd caches Some distributions, notably Fedora, have the following order of nsswitch modules by default: passwd: sss files group: sss files The advantage of serving local users through SSSD is that the nss_sss module has a fast mmapped-cache that speeds up NSS lookups compared to accessing the disk an opening the files on each NSS request. Traditionally, this has been done with the help of nscd, but using nscd in parallel with sssd is cumbersome, as both SSSD and nscd use their own independent caching, so using nscd in setups where sssd is also serving users from some remote domain (LDAP, AD, ...) can result in a bit of unpredictability. More details about why Fedora chose to use sss before files can be found on e.g.: https://fedoraproject.org//wiki/Changes/SSSDCacheForLocalUsers or: https://docs.pagure.org/SSSD.sssd/design_pages/files_provider.html Now, even though sssd watches the passwd and group files with the help of inotify, there can still be a small window where someone requests a user or a group, finds that it doesn't exist, adds the entry and checks again. Without some support in shadow-utils that would explicitly drop the sssd caches, the inotify watch can fire a little late, so a combination of commands like this: getent passwd user || useradd user; getent passwd user can result in the second getent passwd not finding the newly added user as the racy behaviour might still return the cached negative hit from the first getent passwd. This patch more or less copies the already existing support that shadow-utils had for dropping nscd caches, except using the "sss_cache" tool that sssd ships.
2018-09-12 17:52:11 +05:30
sssd_flush_cache (SSSD_DB_GROUP);
return E_SUCCESS;
}
2007-12-28 15:42:09 +05:30