Hash mismatch
esad ergül
Hi Stefano,
We are currently using the SWUpdate Mongoose webserver interface to upload update images. The devices we are using Industrialpcs Intel, AMD, or Celeron processors and are operating with Debian. On these devices, we have an Nginx reverse proxy that forwards requests to nignx swupdate location .
Intermittently, some devices encounter a hash mismatch error during the update process. The update progresses to a certain point, for example, 45% (though this percentage varies each time), and then gets stuck. Subsequently, the update fails and a hash mismatch error is reported. After a few attempts, the update process starts working normally again. However, this error has started to occur more frequently.
We are using Debian Buster and SWUpdate version 2022.05, and we plan to upgrade to Bookworm shortly. Do you have any idea what might be causing this issue?
Below is the Nginx configuration we are using.
location /update/ {
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";
proxy_set_header Connection keep-alive;
proxy_set_header Host $host;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
proxy_cache_bypass $http_upgrade;
proxy_read_timeout 86400;
proxy_send_timeout 86400;
proxy_connect_timeout 86400;
proxy_buffering off;
proxy_request_buffering off;
client_max_body_size 0;
proxy_pass http://localhost:8080/;
}
And the runtime configuration for swupdate is as follows:
webserver :
{
...
timeout = 60;
run-postupdate = false;
};
The error message:
As Info, I generate the SWU update image inside a Bookworm-based Docker container using swugenerator. However, the root filesystem partition image within the SWU is created with ELBE inside a Debian Buster-based container.
Thank you for your help.
Best regards,
Michael Glembotzki
Hi Stefano,
We are currently using the SWUpdate Mongoose webserver interface to upload update images. The devices we are using Industrialpcs Intel, AMD, or Celeron processors and are operating with Debian. On these devices, we have an Nginx reverse proxy that forwards requests to nignx swupdate location .
Intermittently, some devices encounter a hash mismatch error during the update process. The update progresses to a certain point, for example, 45% (though this percentage varies each time), and then gets stuck. Subsequently, the update fails and a hash mismatch error is reported. After a few attempts, the update process starts working normally again. However, this error has started to occur more frequently.
We are using Debian Buster and SWUpdate version 2022.05, and we plan to upgrade to Bookworm shortly. Do you have any idea what might be causing this issue?
Below is the Nginx configuration we are using.
location /update/ {
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";
proxy_set_header Connection keep-alive;
proxy_set_header Host $host;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;
proxy_cache_bypass $http_upgrade;
proxy_read_timeout 86400;
proxy_send_timeout 86400;
proxy_connect_timeout 86400;
proxy_buffering off;
proxy_request_buffering off;
client_max_body_size 0;
Best regards,
Michael
esad ergül
Giorgio Torres (Eugico)
Hi everyone,
Stefano Babic
On 28.08.24 10:58, Giorgio Torres (Eugico) wrote:
>
> Hi everyone,
>
> I'm facing the same issue. When I throttle the network down, let's say
> to 3G, when uploading an swu image, after 1 minute I get this error on
> the server.
> I've already tried to change timeout values for Mongoose but with no
> success.
> I'm wondering if this
It has nothing to do. The timeout is for internal IPC in case a process
establish a connection to command SWUpdate. when a REQ_INSTALL is sent,
the connection is not closed, and the stream is passed to the core:
https://github.com/sbabic/swupdate/blob/32431be199ce408f3280b62b33111fa919c92238/core/network_thread.c#L488
It is on a different path, and the timeout is not set at all.
>
> I'll be following this thread, please if you have any updates or manage
> to bypass this problem, I really appreciate if you let me know 😊
going on. The core has no timeout, because it can take a while to
install a chunk / artifact. The only timeout is in mongoose:
https://github.com/sbabic/swupdate/blob/32431be199ce408f3280b62b33111fa919c92238/mongoose/mongoose_interface.c#L599
Set with :
https://github.com/sbabic/swupdate/blob/32431be199ce408f3280b62b33111fa919c92238/examples/configuration/swupdate.cfg#L274
Exactly what you did.
Best regards,
Stefano Babic
> 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/8f031080-43a8-46af-b62c-02ed082aad40n%40googlegroups.com <https://groups.google.com/d/msgid/swupdate/8f031080-43a8-46af-b62c-02ed082aad40n%40googlegroups.com?utm_medium=email&utm_source=footer>.