TMPDIR changes failing IPC socket creation in yocto

73 views
Skip to first unread message

Meera Shah

unread,
Mar 16, 2023, 2:34:06 AM3/16/23
to swup...@googlegroups.com
Hello,
         I am using swupdate with yocto and in that I am facing issues with /tmp directory's size. As below my /tmp directory size is 431M and my new image size is more than that. That's why I want to redirect my /tmp directory to "/". For that I have also changed socket paths accordingly.

root@oldsilicon:~# df -h
Filesystem            Size  Used Avail Use% Mounted on
/dev/root              15G  871M   13G   7% /
devtmpfs              238M     0  238M   0% /dev
tmpfs                 431M     0  431M   0% /dev/shm
tmpfs                 431M   18M  414M   5% /run
tmpfs                 431M     0  431M   0% /sys/fs/cgroup
tmpfs                 431M     0  431M   0% /tmp
tmpfs                 431M   20K  431M   1% /var/volatile
tmpfs                  87M     0   87M   0% /run/user/0

So, I have changed TMPDIR using this command: export TMPDIR=/
CONFIG_SOCKET_CTRL_PATH="/"
CONFIG_SOCKET_PROGRESS_PATH="/"

Error:
root@oldsilicon:~# swupdate -v -H smarc-rzg2lc:1.0 -i smarc-rzg2lc_1.0.swu
SWUpdate v

Licensed under GPLv2. See source distribution for detailed copyright notices.

[INFO ] : SWUPDATE running :  [print_registered_bootloaders] : Registered bootloaders:
[INFO ] : SWUPDATE running :  [print_registered_bootloaders] :    uboot loaded.
[INFO ] : SWUPDATE running :  [main] : Using default bootloader interface: uboot
[INFO ] : SWUPDATE running :  [main] : Running on smarc-rzg2lc Revision 1.0
[INFO ] : SWUPDATE running :  [print_registered_handlers] : Registered handlers:
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       dummy
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       flash
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       lua
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       raw
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       rawfile
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       rawcopy
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       shellscript
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       preinstall
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       postinstall
[TRACE] : SWUPDATE running :  [listener_create] : creating socket at /
[TRACE] : SWUPDATE running :  [network_initializer] : Main loop daemon
[ERROR] : SWUPDATE failed [0] ERROR : Error creating IPC socket /, exiting.
[TRACE] : SWUPDATE running :  [listener_create] : creating socket at /

Stefano Babic

unread,
Mar 17, 2023, 5:54:37 AM3/17/23
to Meera Shah, swup...@googlegroups.com
On 16.03.23 07:33, Meera Shah wrote:
> Hello,
>          I am using swupdate with yocto and in that I am facing
> issues with /tmp directory's size. As below my /tmp directory size is
> 431M and my new image size is more than that.

Why do you not enable stream mode ("installed-directly") for your
artifacts ? It is thought for these purpose.

> That's why I want to
> redirect my /tmp directory to "/". For that I have also changed socket
> paths accordingly.

This suppose that filesystem is always ok, and swupdate has write access
to it.

>
> root@oldsilicon:~# df -h
> Filesystem            Size  Used Avail Use% Mounted on
> /dev/root              15G  871M   13G   7% /
> devtmpfs              238M     0  238M   0% /dev
> tmpfs                 431M     0  431M   0% /dev/shm
> tmpfs                 431M   18M  414M   5% /run
> tmpfs                 431M     0  431M   0% /sys/fs/cgroup
> tmpfs                 431M     0  431M   0% /tmp
> tmpfs                 431M   20K  431M   1% /var/volatile
> tmpfs                  87M     0   87M   0% /run/user/0
>
> So, I have changed TMPDIR using this command: export TMPDIR=/
> CONFIG_SOCKET_CTRL_PATH="/"
> CONFIG_SOCKET_PROGRESS_PATH="/"

Sockets are always created in TMPDIR and patrh is appended. You wil get
something like "//" in this case.

You just need the export command.
> *[ERROR] : SWUPDATE failed [0] ERROR : Error creating IPC socket /,
> exiting.*
> [TRACE] : SWUPDATE running :  [listener_create] : creating socket at /
>
> --

Best regards,
Stefano Babic


--
=====================================================================
DENX Software Engineering GmbH, Managing Director: Erika Unter
HRB 165235 Munich, Office: Kirchenstr.5, 82194 Groebenzell, Germany
Phone: +49-8142-66989-53 Fax: +49-8142-66989-80 Email: sba...@denx.de
=====================================================================

Meera Shah

unread,
Mar 17, 2023, 6:16:25 AM3/17/23
to Stefano Babic, swup...@googlegroups.com
Hi Stefano,
           Please go through below points I have trialed so far.
  1. I redirected my TMPDIR using this command: root@smarc-rzg2lc:~# export TMPDIR=
   2. I am using renesas(rzg2lc) evk, so my CONFIG_ENV_OFFSET=0xFFFE0000 and         CONFIG_ENV_SIZE=0x20000. Now my /etc/fw_env.config would be
      
/dev/mmcblk0boot1               -0x20000        0x20000. So now I am able to get results on fw_printenv 
and fw_setenv.
   3.  Now I am not getting error related to IPC socket. And my rootfs updated sometime(2/10 trials) after pressing reset button from my device. But I am getting this ext4-fs error.
root@smarc-rzg2lc:~# swupdate -v -H smarc-rzg2lc:1.0 -i smarc-rzg2lc_1.0.swu
SWUpdate v

Licensed under GPLv2. See source distribution for detailed copyright notices.

[TRACE] : SWUPDATE running :  [print_registered_bootloaders] : Registered bootloaders:
[TRACE] : SWUPDATE running :  [print_registered_bootloaders] :    uboot loaded.
[INFO ] : SWUPDATE running :  [main] : Using default bootloader interface: uboot
[INFO ] : SWUPDATE running :  [main] : Running on smarc-rzg2lc Revision 1.0
[INFO ] : SWUPDATE running :  [print_registered_handlers] : Registered handlers:
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       dummy
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       flash
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       lua
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       raw
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       rawfile
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       shellscript
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       preinstall
[INFO ] : SWUPDATE running :  [print_registered_handlers] :       postinstall
[TRACE] : SWUPDATE running :  [network_initializer] : Main loop daemon
[TRACE] : SWUPDATE running :  [listener_create] : creating socket at /swupdateprog
[TRACE] : SWUPDATE running :  [listener_create] : creating socket at /sockinstctrl
[TRACE] : SWUPDATE running :  [network_thread] : Incoming network request: processing...
[INFO ] : SWUPDATE started :  Software Update started !
[TRACE] : SWUPDATE running :  [network_initializer] : Software update started
[TRACE] : SWUPDATE running :  [extract_file_to_tmp] : Found file
[TRACE] : SWUPDATE running :  [extract_file_to_tmp] :       filename sw-description
[TRACE] : SWUPDATE running :  [extract_file_to_tmp] :       size 218
[TRACE] : SWUPDATE running :  [get_common_fields] : Version 1.0.1
[TRACE] : SWUPDATE running :  [parse_hw_compatibility] : Accepted Hw Revision : 1.0
[TRACE] : SWUPDATE running :  [_parse_images] : Found Image: core-image-bsp-old_silicon-20230310112945.rootfs.ext4 in device : /dev/mmcblk1p2 for handler raw
[TRACE] : SWUPDATE running :  [check_hw_compatibility] : Hardware smarc-rzg2lc Revision: 1.0
[TRACE] : SWUPDATE running :  [check_hw_compatibility] : Hardware compatibility verified
[TRACE] : SWUPDATE running :  [extract_files] : Found file
[TRACE] : SWUPDATE running :  [extract_files] :       filename core-image-bsp-old_silicon-20230310112945.rootfs.ext4
[TRACE] : SWUPDATE running :  [extract_files] :       size 493383680 required
[TRACE] : SWUPDATE running :  [extract_padding] : Expecting 388 padding bytes at end-of-file
[TRACE] : SWUPDATE running :  [network_initializer] : Valid image found: copying to FLASH
[INFO ] : SWUPDATE running :  Installation in progress
[TRACE] : SWUPDATE running :  [install_single_image] : Found installer for stream core-image-bsp-old_silicon-20230310112945.rootfs.ext4 raw
[  217.451236] EXT4-fs error (device mmcblk1p2): ext4_validate_block_bitmap:390: comm kworker/u4:1: bg 12: bad block bitmap checksum
[  217.575443] EXT4-fs (mmcblk1p2): Delayed block allocation failed for inode 9566 at logical offset 104244 with max blocks 1024 with error 74
[  217.588080] EXT4-fs (mmcblk1p2): This should not happen!! Data will be lost
[  217.588080]
[  217.607009] EXT4-fs error (device mmcblk1p2): ext4_validate_block_bitmap:390: comm kworker/u4:0: bg 7: bad block bitmap checksum
[  217.753210] EXT4-fs error (device mmcblk1p2) in ext4_free_blocks:5528: Filesystem failed CRC
[  217.784724] EXT4-fs error (device mmcblk1p2): ext4_read_block_bitmap_nowait:483: comm kworker/u4:1: Failed to init block bitmap for group 15: -74
[  217.821067] EXT4-fs error (device mmcblk1p2): ext4_validate_block_bitmap:390: comm kworker/u4:1: bg 16: bad block bitmap checksum
[  217.843073] EXT4-fs error (device mmcblk1p2): ext4_read_block_bitmap_nowait:483: comm kworker/u4:1: Failed to init block bitmap for group 17: -74
[  217.870599] EXT4-fs error (device mmcblk1p2): ext4_read_block_bitmap_nowait:483: comm kworker/u4:1: Failed to init block bitmap for group 18: -74
[  217.899540] EXT4-fs error (device mmcblk1p2): ext4_read_block_bitmap_nowait:483: comm kworker/u4:1: Failed to init block bitmap for group 19: -74
[  217.923986] EXT4-fs error (device mmcblk1p2): ext4_read_block_bitmap_nowait:483: comm kworker/u4:1: Failed to init block bitmap for group 20: -74
[  217.948003] EXT4-fs error (device mmcblk1p2): ext4_read_block_bitmap_nowait:483: comm kworker/u4:1: Failed to init block bitmap for group 21: -74
[  218.010498] EXT4-fs (mmcblk1p2): Delayed block allocation failed for inode 9566 at logical offset 105268 with max blocks 1024 with error 117
[  218.023205] EXT4-fs (mmcblk1p2): This should not happen!! Data will be lost
[  218.023205]
[  219.309899] EXT4-fs (mmcblk1p2): Delayed block allocation failed for inode 9566 at logical offset 106292 with max blocks 1024 with error 117
[  219.322551] EXT4-fs (mmcblk1p2): This should not happen!! Data will be lost
[  219.322551]
[  220.553752] EXT4-fs (mmcblk1p2): Delayed block allocation failed for inode 9566 at logical offset 107316 with max blocks 1024 with error 117
[  220.566380] EXT4-fs (mmcblk1p2): This should not happen!! Data will be lost
[  220.566380]
[  222.702019] EXT4-fs error: 10 callbacks suppressed
[  222.702030] EXT4-fs error (device mmcblk1p2) in ext4_reserve_inode_write:5761: Corrupt filesystem
[  222.861165] EXT4-fs error (device mmcblk1p2): ext4_dirty_inode:5963: inode #262159: comm systemd-timesyn: mark_inode_dirty error
[  224.298268] EXT4-fs error (device mmcblk1p2): ext4_validate_block_bitmap:390: comm kworker/u4:1: bg 112: bad block bitmap checksum
[  224.324573] EXT4-fs error (device mmcblk1p2): ext4_validate_block_bitmap:390: comm kworker/u4:1: bg 32: bad block bitmap checksum
[  224.359487] EXT4-fs (mmcblk1p2): Delayed block allocation failed for inode 9566 at logical offset 108340 with max blocks 1024 with error 74
[  224.372213] EXT4-fs (mmcblk1p2): This should not happen!! Data will be lost
[  224.372213]
[  232.831738] EXT4-fs error (device mmcblk1p2): ext4_validate_block_bitmap:390: comm kworker/u4:1: bg 64: bad block bitmap checksum
[  232.851011] EXT4-fs (mmcblk1p2): Delayed block allocation failed for inode 9566 at logical offset 109364 with max blocks 1024 with error 74
[  232.863595] EXT4-fs (mmcblk1p2): This should not happen!! Data will be lost
[  232.863595]
[  235.399426] EXT4-fs (mmcblk1p2): Delayed block allocation failed for inode 9566 at logical offset 110388 with max blocks 1024 with error 117
[  235.412228] EXT4-fs (mmcblk1p2): This should not happen!! Data will be lost
[  235.412228]
[  237.018031] EXT4-fs error (device mmcblk1p2): ext4_validate_block_bitmap:390: comm kworker/u4:1: bg 96: bad block bitmap checksum
[  237.041676] EXT4-fs (mmcblk1p2): Delayed block allocation failed for inode 9566 at logical offset 111412 with max blocks 1024 with error 74
[  237.054254] EXT4-fs (mmcblk1p2): This should not happen!! Data will be lost
[  237.054254]
[  238.476521] EXT4-fs (mmcblk1p2): Delayed block allocation failed for inode 9566 at logical offset 112436 with max blocks 1024 with error 117
[  238.489384] EXT4-fs (mmcblk1p2): This should not happen!! Data will be lost
[  238.489384]
[  249.743072] EXT4-fs error (device mmcblk1p2): ext4_validate_block_bitmap:390: comm kworker/u4:1: bg 0: bad block bitmap checksum
[  249.806695] EXT4-fs (mmcblk1p2): Delayed block allocation failed for inode 9566 at logical offset 113460 with max blocks 1024 with error 74
[  249.819249] EXT4-fs (mmcblk1p2): This should not happen!! Data will be lost
[  249.819249]
[  250.831874] EXT4-fs (mmcblk1p2): Delayed block allocation failed for inode 9566 at logical offset 114484 with max blocks 1024 with error 117
[  250.844500] EXT4-fs (mmcblk1p2): This should not happen!! Data will be lost




From: Stefano Babic <sba...@denx.de>
Sent: Friday, March 17, 2023 3:24 PM
To: Meera Shah <mee...@glidemtech.com>; swup...@googlegroups.com <swup...@googlegroups.com>
Subject: Re: [swupdate] TMPDIR changes failing IPC socket creation in yocto
 

Stefano Babic

unread,
Mar 17, 2023, 6:21:08 AM3/17/23
to Meera Shah, Stefano Babic, swup...@googlegroups.com
Hi Meera,

On 17.03.23 11:16, Meera Shah wrote:
> Hi Stefano,
>            Please go through below points I have trialed so far.

No, sorry - I can confirm you that setting TMPDIR works and there are no
issue doing this. I have also setupt with dual SWUpdate instances
running on same device and started with different TMPDIR.

But if you want I have to check what is goping on your product, I
provide it just as commercial support.

Best regards,
Stefano Babic

>
> 1. I redirected my TMPDIR using this command: root@smarc-rzg2lc:~#
> ------------------------------------------------------------------------
> *From:* Stefano Babic <sba...@denx.de>
> *Sent:* Friday, March 17, 2023 3:24 PM
> *To:* Meera Shah <mee...@glidemtech.com>; swup...@googlegroups.com
> <swup...@googlegroups.com>
> *Subject:* Re: [swupdate] TMPDIR changes failing IPC socket creation in
> --
> You received this message because you are subscribed to the Google
> Groups "swupdate" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to swupdate+u...@googlegroups.com
> <mailto:swupdate+u...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/swupdate/MAXPR01MB3440DA17286EBA4863F918BEACBD9%40MAXPR01MB3440.INDPRD01.PROD.OUTLOOK.COM <https://groups.google.com/d/msgid/swupdate/MAXPR01MB3440DA17286EBA4863F918BEACBD9%40MAXPR01MB3440.INDPRD01.PROD.OUTLOOK.COM?utm_medium=email&utm_source=footer>.

Meera Shah

unread,
Mar 17, 2023, 6:38:40 AM3/17/23
to Stefano Babic, swup...@googlegroups.com
Hello Stefano,
                I can understand your policy. But I want to attach my sw-description, so maybe we can reach to the cause of this ext4-fs error. Please tell me if you find out anything incorrect with this

software =
{
        version = "1.0.1";

        hardware-compatibility: [ "1.0" ];

        images: (
        {
                        filename = "core-image-bsp-old_silicon-20230310112945.rootfs.ext4";
                        device = "/dev/mmcblk1p2";
        }
    );
}

From: Stefano Babic <sba...@denx.de>
Sent: Friday, March 17, 2023 3:51 PM
To: Meera Shah <mee...@glidemtech.com>; Stefano Babic <sba...@denx.de>; swup...@googlegroups.com <swup...@googlegroups.com>
Subject: Re: [swupdate] TMPDIR changes failing IPC socket creation in yocto
 

Stefano Babic

unread,
Mar 17, 2023, 6:51:26 AM3/17/23
to Meera Shah, Stefano Babic, swup...@googlegroups.com
On 17.03.23 11:38, Meera Shah wrote:
> Hello Stefano,
>                 I can understand your policy. But I want to attach my
> sw-description, so maybe we can reach to the cause of this ext4-fs
> error. Please tell me if you find out anything incorrect with this
>
> software =
> {
>         version = "1.0.1";
>
>         hardware-compatibility: [ "1.0" ];
>
>         images: (
>         {
>                         filename =
> "core-image-bsp-old_silicon-20230310112945.rootfs.ext4";
>                         device = "/dev/mmcblk1p2";
>         }
>     );
> }

sw-description is very simple, and it is ok if the device is corrrect.
This depends on your update concept. Later you should complete it (hash
? signed image ? etc.)

Best regards,
Stefano Babic



> ------------------------------------------------------------------------
> *From:* Stefano Babic <sba...@denx.de>
> *Sent:* Friday, March 17, 2023 3:51 PM
> *To:* Meera Shah <mee...@glidemtech.com>; Stefano Babic
> <sba...@denx.de>; swup...@googlegroups.com <swup...@googlegroups.com>
>> https://groups.google.com/d/msgid/swupdate/MAXPR01MB3440DA17286EBA4863F918BEACBD9%40MAXPR01MB3440.INDPRD01.PROD.OUTLOOK.COM <https://groups.google.com/d/msgid/swupdate/MAXPR01MB3440DA17286EBA4863F918BEACBD9%40MAXPR01MB3440.INDPRD01.PROD.OUTLOOK.COM> <https://groups.google.com/d/msgid/swupdate/MAXPR01MB3440DA17286EBA4863F918BEACBD9%40MAXPR01MB3440.INDPRD01.PROD.OUTLOOK.COM?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/swupdate/MAXPR01MB3440DA17286EBA4863F918BEACBD9%40MAXPR01MB3440.INDPRD01.PROD.OUTLOOK.COM?utm_medium=email&utm_source=footer>>.
>
> --
> =====================================================================
> DENX Software Engineering GmbH,        Managing Director: Erika Unter
> HRB 165235 Munich,   Office: Kirchenstr.5, 82194 Groebenzell, Germany
> Phone: +49-8142-66989-53 Fax: +49-8142-66989-80 Email: sba...@denx.de
> =====================================================================
>

Meera Shah

unread,
Mar 17, 2023, 6:55:59 AM3/17/23
to Stefano Babic, swup...@googlegroups.com
Hi,
     Thanks for your support. It means that my sw-description file is correct. So that is not the suspect point for generating ext4-fs error.

From: Stefano Babic <sba...@denx.de>
Sent: Friday, March 17, 2023 4:21 PM
To: Meera Shah <mee...@glidemtech.com>; Stefano Babic <sba...@denx.de>; swup...@googlegroups.com <swup...@googlegroups.com>
Subject: Re: [swupdate] TMPDIR changes failing IPC socket creation in yocto
 

Meera Shah

unread,
Mar 20, 2023, 4:37:48 AM3/20/23
to Stefano Babic, swup...@googlegroups.com
Hi Stefano,
                I need one help from you. Can we change rootfs of SD card when booting process is from sd card. I meant I am booting my yocto based board from sd card, it has 2 partitions(RZ_FAT and RZ_ext). In RZ_FAT there are kernel and device tree, where in RZ_ext there is rootfs.
Now with the swupdate using sw-description, I want to update rootfs runtime of that SD card without making any of the extra partitions in my board. Can I do that? Because whenever I try to perform this task, my rootfs(RZ_ext) folder deleted itself.

Regards,
Meera


From: Meera Shah <mee...@glidemtech.com>
Sent: Friday, March 17, 2023 4:25 PM
To: Stefano Babic <sba...@denx.de>; swup...@googlegroups.com <swup...@googlegroups.com>

Stefano Babic

unread,
Mar 20, 2023, 5:03:48 AM3/20/23
to Meera Shah, swup...@googlegroups.com
Hi Meera,

On 20.03.23 09:37, Meera Shah wrote:
> Hi Stefano,
>                 I need one help from you. Can we change rootfs of SD
> card when booting process is from sd card. I meant I am booting my yocto
> based board from sd card, it has 2 partitions(RZ_FAT and RZ_ext). In
> RZ_FAT there are kernel and device tree,

This is already a *bad* idea, if rootfs is not encrypted, move them to
rootfs. FAT is not thought to be power-cut safe.

> where in RZ_ext there is rootfs.
> Now with the swupdate using sw-description, I want to update rootfs
> runtime of that SD card without making any of the extra partitions in my
> board. Can I do that?

See the attached picture - this is what you are trying to do, and you
can estimate yourself how this is safe. You are replacing the whole
filesystem as image. Any access to your running filesystem will lead to
crashes or errors, and even not predictable effects. Try to do the same
with a PC, you get the same.

> Because whenever I try to perform this task, my
> rootfs(RZ_ext) folder deleted itself.
>

Nobody is surprised.

Best regards,
Stefano Babic

> Regards,
> Meera
>
> ------------------------------------------------------------------------
> *From:* Meera Shah <mee...@glidemtech.com>
> *Sent:* Friday, March 17, 2023 4:25 PM
> *To:* Stefano Babic <sba...@denx.de>; swup...@googlegroups.com
> <swup...@googlegroups.com>
> *Subject:* Re: [swupdate] TMPDIR changes failing IPC socket creation in
> yocto
> Hi,
>      Thanks for your support. It means that my sw-description file is
> correct. So that is not the suspect point for generating *ext4-fs error*.
> ------------------------------------------------------------------------
> *From:* Stefano Babic <sba...@denx.de>
> *Sent:* Friday, March 17, 2023 4:21 PM
>>> https://groups.google.com/d/msgid/swupdate/MAXPR01MB3440DA17286EBA4863F918BEACBD9%40MAXPR01MB3440.INDPRD01.PROD.OUTLOOK.COM <https://groups.google.com/d/msgid/swupdate/MAXPR01MB3440DA17286EBA4863F918BEACBD9%40MAXPR01MB3440.INDPRD01.PROD.OUTLOOK.COM> <https://groups.google.com/d/msgid/swupdate/MAXPR01MB3440DA17286EBA4863F918BEACBD9%40MAXPR01MB3440.INDPRD01.PROD.OUTLOOK.COM <https://groups.google.com/d/msgid/swupdate/MAXPR01MB3440DA17286EBA4863F918BEACBD9%40MAXPR01MB3440.INDPRD01.PROD.OUTLOOK.COM>> <https://groups.google.com/d/msgid/swupdate/MAXPR01MB3440DA17286EBA4863F918BEACBD9%40MAXPR01MB3440.INDPRD01.PROD.OUTLOOK.COM?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/swupdate/MAXPR01MB3440DA17286EBA4863F918BEACBD9%40MAXPR01MB3440.INDPRD01.PROD.OUTLOOK.COM?utm_medium=email&utm_source=footer <https://groups.google.com/d/msgid/swupdate/MAXPR01MB3440DA17286EBA4863F918BEACBD9%40MAXPR01MB3440.INDPRD01.PROD.OUTLOOK.COM?utm_medium=email&utm_source=footer>>>.
>>
>> --
>> =====================================================================
>> DENX Software Engineering GmbH,        Managing Director: Erika Unter
>> HRB 165235 Munich,   Office: Kirchenstr.5, 82194 Groebenzell, Germany
>> Phone: +49-8142-66989-53 Fax: +49-8142-66989-80 Email: sba...@denx.de
>> =====================================================================
>>
>
> --
> =====================================================================
> DENX Software Engineering GmbH,        Managing Director: Erika Unter
> HRB 165235 Munich,   Office: Kirchenstr.5, 82194 Groebenzell, Germany
> Phone: +49-8142-66989-53 Fax: +49-8142-66989-80 Email: sba...@denx.de
> =====================================================================
>
> --
> You received this message because you are subscribed to the Google
> Groups "swupdate" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to swupdate+u...@googlegroups.com
> <mailto:swupdate+u...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/swupdate/MAXPR01MB34402A61A8AACA013B2AF752AC809%40MAXPR01MB3440.INDPRD01.PROD.OUTLOOK.COM <https://groups.google.com/d/msgid/swupdate/MAXPR01MB34402A61A8AACA013B2AF752AC809%40MAXPR01MB3440.INDPRD01.PROD.OUTLOOK.COM?utm_medium=email&utm_source=footer>.
images.jpeg

Meera Shah

unread,
Mar 29, 2023, 5:14:10 AM3/29/23
to Stefano Babic, swup...@googlegroups.com
Hi Stefano,
            Now I can understand what you were trying to say. So, I changed my approach for rootfs update.
There are basically 3 partitions:
(1)RZ_FAT --> 500MB (/dev/mmcblk1p1)
(1)RZ_ext  --> 13.72GiB (/dev/mmcblk1p2) ---> current rootfs partition
(1)RZ_ext2 --> 14.65GiB (/dev/mmcblk1p3) ---> after swupdate rootfs partition

Here is my  sw-description and update.sh script. Right now I am booting up my rootfs from /dev/mmcblk1p2, and In sw-description I am giving /dev/mmcblk1p3 for rootfs.
And I can successfully update rootfs. But as soon as, I updated rootfs my partition(/dev/mmcblk1p3) size reduced automatically.

root@smarc-rzg2lc:~# df -h 

Filesystem            Size  Used Avail Use% Mounted on
/dev/root             667M  558M   59M  91% /
devtmpfs              738M     0  738M   0% /dev
tmpfs                 932M     0  932M   0% /dev/shm
tmpfs                 932M   18M  914M   2% /run
tmpfs                 932M     0  932M   0% /sys/fs/cgroup
tmpfs                 932M     0  932M   0% /tmp
tmpfs                 932M   24K  932M   1% /var/volatile
tmpfs                 187M     0  187M   0% /run/user/0




From: Stefano Babic <sba...@denx.de>
Sent: Monday, March 20, 2023 2:33 PM
To: Meera Shah <mee...@glidemtech.com>; swup...@googlegroups.com <swup...@googlegroups.com>
Subject: Re: [swupdate] TMPDIR changes failing IPC socket creation in yocto
 
Reply all
Reply to author
Forward
0 new messages