Re: [PATCH BlueZ] mesh: Log D-Bus method call errors

From: "Gix, Brian" <brian.gix@intel.com>
To: "michal.lowas-rzechonek@silvair.com" 
	<michal.lowas-rzechonek@silvair.com>
Cc: "linux-bluetooth@vger.kernel.org" <linux-bluetooth@vger.kernel.org>
Subject: Re: [PATCH BlueZ] mesh: Log D-Bus method call errors
Date: Thu, 29 Aug 2019 20:07:26 +0000	[thread overview]
Message-ID: <38E88E55-4291-47AF-B502-47922E9FBA56@intel.com> (raw)
In-Reply-To: <20190829195610.a6dwgxabq3d2g3bp@kynes>

Hi Michał,

> On Aug 29, 2019, at 12:56 PM, "michal.lowas-rzechonek@silvair.com" <michal.lowas-rzechonek@silvair.com> wrote:
> 
> Hi Brian,
> 
>> On 08/29, Gix, Brian wrote:
>> That is true, and we *expect* applications that are attached to handle
>> the socket-signals (that drive d-bus) in a timely manner...  But I am
>> not sure we have a way to enforce it.
>> 
>> Certainly, we can simulate a disconnection if an App ignores it's DBus
>> socket signal, but again, we won't know about that for 30 seconds
>> which seems like forever...  And an App could potentially have a large
>> enough backlog of messages negatively affecting the daemon before it
>> and corrects it.
> 
> This seems like a limitation of ELL:
> 1. There doesn't seem to be an explicit API to set timeouts on method
>   calls, so if the application takes too long to handle method calls,
>   message_list hashmap in l_dbus struct would indeed grow quite large.
> 2. There doesn't seem to be a way to set an upper limit of pending
>   messages (or maybe even message rate?) in l_dbus connection.
> 
> Still, looking at ell/dbus.c, it seems it should be possible to manually
> call l_dbus_cancel after obtaining serial number of the method call
> message, using _dbus_message_get_serial from dbus-private.h (yeah, I
> know).
> 
> Anyway, I think a better approach would be to submit patches to ELL
> implementing these two features, and then use these additions in meshd.
> Does that sound acceptable from your POV?

I’m not sure I agree with you on the need to expose and adjust  DBus message timeouts, however, I think you are probably correct that the correct place to have that conversation is on the ELL reflector, which can be accessed here:

https://lists.01.org/mailman/listinfo/ell

I still feel though, that you are trying to solve a pre-deployment “debugging” issue by requiring DBus responses for messages that don’t require them.

> 
> By the way, it seems that bluetoothd suffers from the same problem with
> regards to external GATT services/characteristics/descriptors.
> 
> regards
> -- 
> Michał Lowas-Rzechonek <michal.lowas-rzechonek@silvair.com>
> Silvair http://silvair.com
> Jasnogórska 44, 31-358 Krakow, POLAND