[PATCH] huawei: retry connect/disconnect attempt upon NDISSTATQRY failures

[Ben Chan]

On Mon, Sep 2, 2013 at 10:44 AM, Aleksander Morgado
<aleksander at lanedo.com> wrote:
> On 02/09/13 19:18, Ben Chan wrote:
>>     See some comments below.
>>
>>     But I'm mostly interested in what the exact issues are. What does
>>     NDISSTATQRY reply that the parser doesn't like? Which is the exact
>>     response string? Shouldn't we just improve the parser to handle
>>     those cases?
>>
>>
>> We observe an issue on MU736 that ^NDISSTATQRY? sometimes reports +CME
>> ERROR: 100 (Unknown error). The modem recovers if we retry the command,
>> which is what this patch tries to accomplish.
>>
>
> I see. A comment explaining that in the patch itself is always
> appreciated :)
>
>> BTW, should we use the "^NDISSTAT" unsolicited message to determine the
>> connection status, instead of polling via "^NDISSTATQRY?" ?
>>
>
> Definitely. If NDISSTAT unsolicited messages give us this information
> without needing to poll, and they are more reliable, then we should use
> them, as we do e.g. in the option/hso plugin.
>

Aleksander,  I've updated Prathmesh's patch based on your suggestions.

Thanks,
Ben