srp_daemon: Avoid extra permissions for the lock file
PATCH 1/1 srp_daemon: Avoid extra permissions for the lock file
There is no need to create a world-writable lock file. It's enough to have an RW permission for the file owner only.
--- srp_daemon/srp_daemon.c | 1 - 1 file changed, 1 deletion(-)
srp_daemon/srp_daemon.c +0 -1
@@ -142,7 +142,6 @@ static int check_process_uniqueness(struct config_t *conf)return -1 ; } -fchmod (fd,S_IRUSR |S_IRGRP |S_IROTH |S_IWUSR |S_IWGRP |S_IWOTH );if (0 !=lockf (fd,F_TLOCK ,0 )) {pr_err ("failed to lock %s (errno: %d). possibly another " "srp_daemon is locking it\n" , path, errno);
Signature · 1 line
2.21.1
PATCH 1/1 srp_daemon: Avoid extra permissions for the lock file
Quoted 31 lines
From: Sergey Gorenko <[email protected]>
Sent: Wednesday, August 19, 2020 5:18 PM
To: [email protected]
Cc: [email protected]; Sergey Gorenko <[email protected]>; Max Gurtovoy <[email protected]>
Subject: [PATCH]srp_daemon: Avoid extra permissions for the lock fileThere is no need to create a world-writable lock file. It's enough to have an RW permission for the file owner only.
Signed-off-by Sergey Gorenko <[email protected]>Reviewed-by Max Gurtovoy <[email protected]>--- srp_daemon/srp_daemon.c | 1 - 1 file changed, 1 deletion(-)srp_daemon/srp_daemon.c index renamed from srp_daemon/srp_daemon.c +0
-0
f14d9f56c9f2..fcf94537cebb100644srp_daemon/srp_daemon.c +0
-1@@ -142,7 +142,6 @@ static int check_process_uniqueness(struct config_t *conf)return -1 ; } -fchmod (fd, S_IRUSR|S_IRGRP|S_IROTH|S_IWUSR|S_IWGRP|S_IWOTH);if (0 !=lockf (fd,F_TLOCK ,0 )) {srp_daemon/srp_daemon.c +0
-1pr_err ("failed to lock %s (errno: %d). possibly another " "srp_daemon is locking it\n" , path, errno); --
Quoted 1 line
2.21.1
Hi Bart,
Could you review the patch? I'm asking for you because you are specified as a maintainer for srp_daemon in rdma-core.
Regards,
Sergey
Quoted 2 lines · Sergey Gorenko
Could you review the patch? I'm asking for you because you are specified as a maintainer for
srp_daemonin rdma-core.
Thanks for the reminder. That patch had escaped from my attention.
Bart.
Quoted 14 lines · Sergey Gorenko
There is no need to create a world-writable lock file. It's enough to have an RW permission for the file owner only.
Signed-off-by Sergey Gorenko <[email protected]>Reviewed-by Max Gurtovoy <[email protected]>--- srp_daemon/srp_daemon.c | 1 - 1 file changed, 1 deletion(-)srp_daemon/srp_daemon.c +0
-0
srp_daemon/srp_daemon.c static int check_process_uniqueness(struct config_t *conf) · line 147
Quoted 7 lines
srp_daemon/srp_daemon.c +0
-1@@ -142,7 +142,6 @@ static int check_process_uniqueness(struct config_t *conf)return -1 ; } -fchmod (fd,S_IRUSR |S_IRGRP |S_IROTH |S_IWUSR |S_IWGRP |S_IWOTH );if (0 !=lockf (fd,F_TLOCK ,0 )) {pr_err ("failed to lock %s (errno: %d). possibly another " "srp_daemon is locking it\n" , path, errno);
I think the fchmod() call was introduced by commit ee138ce1e40d ("Cause srp_daemon launch to fail if another srp_daemon is already working on the same HCA port."). Has it been verified that with this change applied that mechanism still works?
Anyway, please add a reference to that commit in the patch description.
Thanks,
Bart.
Quoted 38 lines
-----Original Message-----
From: Bart Van Assche <[email protected]>
Sent: Monday, September 14, 2020 7:56 PM
To: Sergey Gorenko <[email protected]>
Cc: [email protected]; Max Gurtovoy <[email protected]>
Subject: Re: [PATCH]srp_daemon: Avoid extra permissions for the lock filesrp_daemon/srp_daemon.c static int check_process_uniqueness(struct
Quoted 14 lines · Sergey Gorenko
There is no need to create a world-writable lock file. It's enough to have an RW permission for the file owner only.
Signed-off-by Sergey Gorenko <[email protected]>Reviewed-by Max Gurtovoy <[email protected]>--- srp_daemon/srp_daemon.c | 1 - 1 file changed, 1 deletion(-)srp_daemon/srp_daemon.c index renamed from srp_daemon/srp_daemon.c +0
-0
f14d9f56c9f2..fcf94537cebb100644srp_daemon/srp_daemon.c +0
-0@@ -142,7 +142,6 @@ static int check_process_uniqueness(struct
config_t*conf)srp_daemon/srp_daemon.c static int check_process_uniqueness(struct config_t *conf)
Quoted 4 lines
srp_daemon/srp_daemon.c +0
-1return -1 ; } - fchmod(fd,S_IRUSR|S_IRGRP|S_IROTH|S_IWUSR|S_IWGRP|S_IWOTH);
srp_daemon/srp_daemon.c static int check_process_uniqueness(struct config_t *conf) · line 147
Quoted 3 lines
srp_daemon/srp_daemon.c +0
-0if (0 !=lockf (fd,F_TLOCK ,0 )) {pr_err ("failed to lock %s (errno: %d). possibly another " "srp_daemon is locking it\n" , path, errno);I think the
fchmod()call was introduced by commitee138ce1e40d("Causesrp_daemonlaunch to fail if anothersrp_daemonis already working on the same HCA port."). Has it been verified that with this change applied that mechanism still works?
Quoted 6 lines
Anyway, please add a reference to that commit in the patch description.
Thanks,
Bart.
Bart,
I tested the patch for the following scenarios:
* Start the srp_daemon service when srp_daemon is not running and the lock file does not exist.
* Start the srp_daemon service when srp_daemon is not running and the lock file exists.
* Start the srp_daemon service when srp_daemon is running and the lock file exists.
* Start the srp_daemon service when srp_daemon is running and the lock file exists and the file owner is not root. (Such scenario can happen if someone tries to run srp_daemon manually as not root. The srp_daemon fails in this case, but the lock file is created). This case is handled successfully even without the fchmod() call because the srp_daemon service starts srp_daemon as root.
I do not know any case when fchmod() is needed. And it does not look like a good idea to create a word-writable file owned by root. That's why I want to remove the fchmod() call.
Do you have an idea when the fchmod() call can be needed?
If you have no other objections, I will add the fixes line and send V1.
Thanks,
Sergey
Bart Van Assche
Reviewed-byQuoted 11 lines · Sergey Gorenko
I tested the patch for the following scenarios:
* Start thesrp_daemonservice whensrp_daemonis not running and the lock file does not exist.
* Start thesrp_daemonservice whensrp_daemonis not running and the lock file exists.
* Start thesrp_daemonservice whensrp_daemonis running and the lock file exists.
* Start thesrp_daemonservice whensrp_daemonis running and the lock file exists and the file owner is not root. (Such scenario can happen if someone tries to runsrp_daemonmanually as not root. Thesrp_daemonfails in this case, but the lock file is created). This case is handled successfully even without thefchmod()call because thesrp_daemonservice startssrp_daemonas root.I do not know any case when
fchmod()is needed. And it does not look like a good idea to create a word-writable file owned by root. That's why I want to remove thefchmod()call.Do you have an idea when the
fchmod()call can be needed?If you have no other objections, I will add the fixes line and send V1.
Thanks Sergey for having shared all this information. I think this testing is sufficient. Hence: