sstate: fix touching files inside pseudo
PATCH v2 1/1 sstate: fix touching files inside pseudo
running the 'id' command inside the sstate_create_package function shows that this funcion run inside the pseudo:
uid=0(root) gid=0(root) groups=0(root)
The check for touch files [ ! -w ${SSTATE_PKG} ] will always return true and the touch can fail when the real user don't have permission or
in readonly filesystem.
As the documentation refers, the file test operator "-w" check if the file has write permission (for the user running the test).
We can avoid this test running the touch and mask any return errors that we have.
---
v2: redirect stderr to /dev/null to avoid messages in the log
meta/classes/sstate.bbclass | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
meta/classes/sstate.bbclass +2 -2
@@ -830,7 +830,7 @@ sstate_task_postfunc[dirs] = "${WORKDIR}" sstate_create_package () { # Exit early if it already exists if [ -e ${SSTATE_PKG} ]; then - [ ! -w ${SSTATE_PKG} ] || touch ${SSTATE_PKG} + touch ${SSTATE_PKG} 2>/dev/null || true return fi
@@ -865,7 +865,7 @@ sstate_create_package () { else rm $TFILE fi - [ ! -w ${SSTATE_PKG} ] || touch ${SSTATE_PKG} + touch ${SSTATE_PKG} 2>/dev/null || true } python sstate_sign_package () {
Signature · 1 line
2.33.1
Quoted 55 lines
-----Original Message-----
From: [email protected] <openembedded- [email protected]> On Behalf Of Jose Quaresma
Sent: den 20 oktober 2021 19:26
To: [email protected]
Cc: Jose Quaresma <[email protected]>
Subject: [OE-core] [PATCH v2] sstate: fix touching files inside pseudorunning the 'id' command inside the
sstate_create_packagefunction shows that this funcion run inside the pseudo:uid=0(root) gid=0(root) groups=0(root)
The check for touch files [ ! -w ${SSTATE_PKG} ] will always return true and the touch can fail when the real user don't have permission or
in readonly filesystem.As the documentation refers, the file test operator "-w" check if the file has write permission (for the user running the test).
We can avoid this test running the touch and mask any return errors that we have.
Signed-off-by Jose Quaresma <[email protected]>---
v2: redirect stderr to
/dev/nullto avoid messages in the logmeta/classes/sstate.bbclass | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)meta/classes/sstate.bbclass +2
-3@@ -830,7 +830,7 @@ sstate_task_postfunc[dirs] = "${WORKDIR}" sstate_create_package () { # Exit early if it already exists if [ -e ${SSTATE_PKG} ]; then - [ ! -w ${SSTATE_PKG} ] || touch ${SSTATE_PKG} + touch ${SSTATE_PKG} 2>/dev/null || true return fi@@ -865,7 +865,7 @@ sstate_create_package () { else rm $TFILE fi - [ ! -w ${SSTATE_PKG} ] || touch ${SSTATE_PKG} + touch ${SSTATE_PKG} 2>/dev/null || true } python sstate_sign_package () { --
Quoted 1 line
2.33.1
Can this please be backported to Hardknott and Honister now that it has been accepted for master?
//Peter
I have found this bug on the dunfell branch, can it be backported to dunfell too?
Jose
Peter Kjellerstedt <[email protected]> escreveu no dia segunda, 25/10/2021 à(s) 15:28:
Quoted 63 lines
Quoted 55 lines
-----Original Message-----
From: [email protected] <openembedded- [email protected]> On Behalf Of Jose Quaresma
Sent: den 20 oktober 2021 19:26
To: [email protected]
Cc: Jose Quaresma <[email protected]>
Subject: [OE-core] [PATCH v2] sstate: fix touching files inside pseudorunning the 'id' command inside the
sstate_create_packagefunction shows that this funcion run inside the pseudo:uid=0(root) gid=0(root) groups=0(root)
The check for touch files [ ! -w ${SSTATE_PKG} ] will always return true and the touch can fail when the real user don't have permission or
in readonly filesystem.As the documentation refers, the file test operator "-w" check if the file has write permission (for the user running the test).
We can avoid this test running the touch and mask any return errors that we have.
Signed-off-by Jose Quaresma <[email protected]>---
v2: redirect stderr to
/dev/nullto avoid messages in the logmeta/classes/sstate.bbclass | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)meta/classes/sstate.bbclass +2
-3@@ -830,7 +830,7 @@ sstate_task_postfunc[dirs] = "${WORKDIR}" sstate_create_package () { # Exit early if it already exists if [ -e ${SSTATE_PKG} ]; then - [ ! -w ${SSTATE_PKG} ] || touch ${SSTATE_PKG} + touch ${SSTATE_PKG} 2>/dev/null || true return fi@@ -865,7 +865,7 @@ sstate_create_package () { else rm $TFILE fi - [ ! -w ${SSTATE_PKG} ] || touch ${SSTATE_PKG} + touch ${SSTATE_PKG} 2>/dev/null || true } python sstate_sign_package () { --Quoted 1 line
2.33.1
Can this please be backported to Hardknott and Honister now that it has been accepted for master?
//Peter
Signature · 3 lines
Best regards, José Quaresma
Quoted 2 lines · Jose Quaresma
I have found this bug on the dunfell branch, can it be backported to dunfell too?
I'll add it to my testing queue.
Steve
Quoted 82 lines
Jose
Peter Kjellerstedt <[email protected]> escreveu no dia segunda, 25/10/2021 à(s) 15:28:
Quoted 54 lines
Quoted 54 lines
-----Original Message-----
From: [email protected] <openembedded- [email protected]> On Behalf Of Jose Quaresma
Sent: den 20 oktober 2021 19:26
To: [email protected]
Cc: Jose Quaresma <[email protected]>
Subject: [OE-core] [PATCH v2] sstate: fix touching files inside pseudorunning the 'id' command inside the
sstate_create_packagefunction shows that this funcion run inside the pseudo:uid=0(root) gid=0(root) groups=0(root)
The check for touch files [ ! -w ${SSTATE_PKG} ] will always return true and the touch can fail when the real user don't have permission or
in readonly filesystem.As the documentation refers, the file test operator "-w" check if the file has write permission (for the user running the test).
We can avoid this test running the touch and mask any return errors that we have.
Signed-off-by Jose Quaresma <[email protected]>---
v2: redirect stderr to
/dev/nullto avoid messages in the logmeta/classes/sstate.bbclass | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)meta/classes/sstate.bbclass +2
-2@@ -830,7 +830,7 @@ sstate_task_postfunc[dirs] = "${WORKDIR}" sstate_create_package () { # Exit early if it already exists if [ -e ${SSTATE_PKG} ]; then - [ ! -w ${SSTATE_PKG} ] || touch ${SSTATE_PKG} + touch ${SSTATE_PKG} 2>/dev/null || true return fi@@ -865,7 +865,7 @@ sstate_create_package () { else rm $TFILE fi - [ ! -w ${SSTATE_PKG} ] || touch ${SSTATE_PKG} + touch ${SSTATE_PKG} 2>/dev/null || true }Quoted 9 lines
Quoted 3 lines
meta/classes/sstate.bbclass +0
-1python sstate_sign_package () { --2.33.1
Can this please be backported to Hardknott and Honister now that it has been accepted for master?
//Peter
--
Best regards,José Quaresma
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#157377): https://lists.openembedded.org/g/openembedded-core/message/157377
Mute This Topic: https://lists.openembedded.org/mt/86470820/3617601
Group Owner: [email protected]
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub [[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-