[RFC] gatchat: Print error message if opening tun failes

[RFC] gatchat: Print error message if opening tun failes

[Daniel Wagner]

[-- Attachment #1: Type: text/plain, Size: 824 bytes --]

From: Daniel Wagner <daniel.wagner@bmw-carit.de>

This is a very common misstake. Let's help the users to
configure their system correctly.
---
 gatchat/ppp_net.c | 6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)

diff --git a/gatchat/ppp_net.c b/gatchat/ppp_net.c
index 1609b99..7082de7 100644
--- a/gatchat/ppp_net.c
+++ b/gatchat/ppp_net.c
@@ -155,8 +155,12 @@ struct ppp_net *ppp_net_new(GAtPPP *ppp, int fd)
 	if (fd < 0) {
 		/* open a tun interface */
 		fd = open("/dev/net/tun", O_RDWR);
-		if (fd < 0)
+		if (fd < 0) {
+			fprintf(stderr, "Couldn't open tun device. "
+					"Do you run oFono as root and do you "
+					"have the TUN module loaded?");
 			goto error;
+		}
 
 		ifr.ifr_flags = IFF_TUN | IFF_NO_PI;
 		strcpy(ifr.ifr_name, "ppp%d");
-- 
1.7.12.rc1.16.g05a20c8

Re: [RFC] gatchat: Print error message if opening tun failes

[Marcel Holtmann]

[-- Attachment #1: Type: text/plain, Size: 904 bytes --]

Hi Daniel,

> This is a very common misstake. Let's help the users to
> configure their system correctly.
> ---
>  gatchat/ppp_net.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> diff --git a/gatchat/ppp_net.c b/gatchat/ppp_net.c
> index 1609b99..7082de7 100644
> --- a/gatchat/ppp_net.c
> +++ b/gatchat/ppp_net.c
> @@ -155,8 +155,12 @@ struct ppp_net *ppp_net_new(GAtPPP *ppp, int fd)
>  	if (fd < 0) {
>  		/* open a tun interface */
>  		fd = open("/dev/net/tun", O_RDWR);
> -		if (fd < 0)
> +		if (fd < 0) {
> +			fprintf(stderr, "Couldn't open tun device. "
> +					"Do you run oFono as root and do you "
> +					"have the TUN module loaded?");

really, stderr?

>  			goto error;
> +		}
>  
>  		ifr.ifr_flags = IFF_TUN | IFF_NO_PI;
>  		strcpy(ifr.ifr_name, "ppp%d");

Aren't the atom drivers checking this case for you?

Regards

Marcel

Re: [RFC] gatchat: Print error message if opening tun failes

[Daniel Wagner]

[-- Attachment #1: Type: text/plain, Size: 1289 bytes --]

Hi Marcel,

On 21.08.2012 15:00, Marcel Holtmann wrote:
>> This is a very common misstake. Let's help the users to
>> configure their system correctly.
>> ---
>>   gatchat/ppp_net.c | 6 +++++-
>>   1 file changed, 5 insertions(+), 1 deletion(-)
>>
>> diff --git a/gatchat/ppp_net.c b/gatchat/ppp_net.c
>> index 1609b99..7082de7 100644
>> --- a/gatchat/ppp_net.c
>> +++ b/gatchat/ppp_net.c
>> @@ -155,8 +155,12 @@ struct ppp_net *ppp_net_new(GAtPPP *ppp, int fd)
>>   	if (fd < 0) {
>>   		/* open a tun interface */
>>   		fd = open("/dev/net/tun", O_RDWR);
>> -		if (fd < 0)
>> +		if (fd < 0) {
>> +			fprintf(stderr, "Couldn't open tun device. "
>> +					"Do you run oFono as root and do you "
>> +					"have the TUN module loaded?");
>
> really, stderr?

I was not sure what 'channel' to use to report something from gatchat. 
Would g_print() be the right choice?

>>   			goto error;
>> +		}
>>
>>   		ifr.ifr_flags = IFF_TUN | IFF_NO_PI;
>>   		strcpy(ifr.ifr_name, "ppp%d");
>
> Aren't the atom drivers checking this case for you?

I don't know, but I had the exactly the same problem and I am pretty 
sure that there was not log entry in any kind. If there is better place 
to add such a log entry, I happy to send a patch.

cheers,
daniel

Re: [RFC] gatchat: Print error message if opening tun failes

[Daniel Wagner]

[-- Attachment #1: Type: text/plain, Size: 1353 bytes --]

Hi Marcel,

[sorry, a resend, ofono doesn't like non subscriber to post]

On 21.08.2012 15:00, Marcel Holtmann wrote:
>> This is a very common misstake. Let's help the users to
>> configure their system correctly.
>> ---
>>   gatchat/ppp_net.c | 6 +++++-
>>   1 file changed, 5 insertions(+), 1 deletion(-)
>>
>> diff --git a/gatchat/ppp_net.c b/gatchat/ppp_net.c
>> index 1609b99..7082de7 100644
>> --- a/gatchat/ppp_net.c
>> +++ b/gatchat/ppp_net.c
>> @@ -155,8 +155,12 @@ struct ppp_net *ppp_net_new(GAtPPP *ppp, int fd)
>>   	if (fd < 0) {
>>   		/* open a tun interface */
>>   		fd = open("/dev/net/tun", O_RDWR);
>> -		if (fd < 0)
>> +		if (fd < 0) {
>> +			fprintf(stderr, "Couldn't open tun device. "
>> +					"Do you run oFono as root and do you "
>> +					"have the TUN module loaded?");
>
> really, stderr?

I was not sure what 'channel' to use to report something from gatchat. 
Would g_print() be the right choice?

>>   			goto error;
>> +		}
>>
>>   		ifr.ifr_flags = IFF_TUN | IFF_NO_PI;
>>   		strcpy(ifr.ifr_name, "ppp%d");
>
> Aren't the atom drivers checking this case for you?

I don't know, but I had the exactly the same problem and I am pretty 
sure that there was not log entry in any kind. If there is better place 
to add such a log entry, I happy to send a patch.

cheers,
daniel

Re: [RFC] gatchat: Print error message if opening tun failes

[Denis Kenzior]

[-- Attachment #1: Type: text/plain, Size: 1589 bytes --]

Hi Daniel,

On 08/21/2012 10:09 AM, Daniel Wagner wrote:
> Hi Marcel,
>
> [sorry, a resend, ofono doesn't like non subscriber to post]
>
> On 21.08.2012 15:00, Marcel Holtmann wrote:
>>> This is a very common misstake. Let's help the users to
>>> configure their system correctly.
>>> ---
>>> gatchat/ppp_net.c | 6 +++++-
>>> 1 file changed, 5 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/gatchat/ppp_net.c b/gatchat/ppp_net.c
>>> index 1609b99..7082de7 100644
>>> --- a/gatchat/ppp_net.c
>>> +++ b/gatchat/ppp_net.c
>>> @@ -155,8 +155,12 @@ struct ppp_net *ppp_net_new(GAtPPP *ppp, int fd)
>>> if (fd < 0) {
>>> /* open a tun interface */
>>> fd = open("/dev/net/tun", O_RDWR);
>>> - if (fd < 0)
>>> + if (fd < 0) {
>>> + fprintf(stderr, "Couldn't open tun device. "
>>> + "Do you run oFono as root and do you "
>>> + "have the TUN module loaded?");
>>
>> really, stderr?
>
> I was not sure what 'channel' to use to report something from gatchat.
> Would g_print() be the right choice?
>

ppp_debug would be best

>>> goto error;
>>> + }
>>>
>>> ifr.ifr_flags = IFF_TUN | IFF_NO_PI;
>>> strcpy(ifr.ifr_name, "ppp%d");
>>
>> Aren't the atom drivers checking this case for you?
>
> I don't know, but I had the exactly the same problem and I am pretty
> sure that there was not log entry in any kind. If there is better place
> to add such a log entry, I happy to send a patch.
>

There is a check for TUN inside drivers/atmodem/gprs-context.c 
at_gprs_context_probe().  Why it isn't being triggered I do not know.

Regards,
-Denis