[PATCH] cryptroot-ask: check that plymouth is running before use

[PATCH] cryptroot-ask: check that plymouth is running before use

[Andrey Borzenkov]

Otherwise there is no way to skip pasword prompt. --has-active-vt
seems to correctly catch also the case when plymouthd is started
but splash is disabled.

Signed-off-by: Andrey Borzenkov <arvidjaar-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>

---
 modules.d/90crypt/cryptroot-ask.sh |    5 ++---
 1 files changed, 2 insertions(+), 3 deletions(-)

diff --git a/modules.d/90crypt/cryptroot-ask.sh b/modules.d/90crypt/cryptroot-ask.sh
index 5150688..9c99ccc 100755
--- a/modules.d/90crypt/cryptroot-ask.sh
+++ b/modules.d/90crypt/cryptroot-ask.sh
@@ -80,9 +80,8 @@ if [ -n "$(getarg rd.luks.key)" ]; then
     rmdir "$mntp"
     unset mntp keypath keydev
 else
-    # Prompt for password with plymouth, if installed.
-    # Should we check if plymouthd is running?
-    if [ -x /bin/plymouth ]; then
+    # Prompt for password with plymouth, if installed and running.
+    if [ -x /bin/plymouth ] && /bin/plymouth --has-active-vt; then
         prompt="Password [$device ($luksname)]:" 
         if [ ${#luksname} -gt 8 ]; then
             sluksname=${sluksname##luks-}
-- 
tg: (661f9a3..) upstream/crypto-plymouth (depends on: origin/master)

Re: [PATCH] cryptroot-ask: check that plymouth is running before use

[Harald Hoyer]

Am 10.03.2011 23:54, schrieb Andrey Borzenkov:
> Otherwise there is no way to skip pasword prompt. --has-active-vt
> seems to correctly catch also the case when plymouthd is started
> but splash is disabled.
> 
> Signed-off-by: Andrey Borzenkov <arvidjaar-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
> 
> ---
>  modules.d/90crypt/cryptroot-ask.sh |    5 ++---
>  1 files changed, 2 insertions(+), 3 deletions(-)
> 
> diff --git a/modules.d/90crypt/cryptroot-ask.sh b/modules.d/90crypt/cryptroot-ask.sh
> index 5150688..9c99ccc 100755
> --- a/modules.d/90crypt/cryptroot-ask.sh
> +++ b/modules.d/90crypt/cryptroot-ask.sh
> @@ -80,9 +80,8 @@ if [ -n "$(getarg rd.luks.key)" ]; then
>      rmdir "$mntp"
>      unset mntp keypath keydev
>  else
> -    # Prompt for password with plymouth, if installed.
> -    # Should we check if plymouthd is running?
> -    if [ -x /bin/plymouth ]; then
> +    # Prompt for password with plymouth, if installed and running.
> +    if [ -x /bin/plymouth ] && /bin/plymouth --has-active-vt; then
>          prompt="Password [$device ($luksname)]:" 
>          if [ ${#luksname} -gt 8 ]; then
>              sluksname=${sluksname##luks-}

Very nice! Thank you!

Re: [PATCH] cryptroot-ask: check that plymouth is running before use

[Andrey Borzenkov]

[-- Attachment #1: Type: text/plain, Size: 1451 bytes --]

On Fri, Mar 11, 2011 at 11:41 AM, Harald Hoyer <harald-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> wrote:
> Am 10.03.2011 23:54, schrieb Andrey Borzenkov:
>> Otherwise there is no way to skip pasword prompt. --has-active-vt
>> seems to correctly catch also the case when plymouthd is started
>> but splash is disabled.
>>
>> Signed-off-by: Andrey Borzenkov <arvidjaar-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
>>
>> ---
>>  modules.d/90crypt/cryptroot-ask.sh |    5 ++---
>>  1 files changed, 2 insertions(+), 3 deletions(-)
>>
>> diff --git a/modules.d/90crypt/cryptroot-ask.sh b/modules.d/90crypt/cryptroot-ask.sh
>> index 5150688..9c99ccc 100755
>> --- a/modules.d/90crypt/cryptroot-ask.sh
>> +++ b/modules.d/90crypt/cryptroot-ask.sh
>> @@ -80,9 +80,8 @@ if [ -n "$(getarg rd.luks.key)" ]; then
>>      rmdir "$mntp"
>>      unset mntp keypath keydev
>>  else
>> -    # Prompt for password with plymouth, if installed.
>> -    # Should we check if plymouthd is running?
>> -    if [ -x /bin/plymouth ]; then
>> +    # Prompt for password with plymouth, if installed and running.
>> +    if [ -x /bin/plymouth ] && /bin/plymouth --has-active-vt; then
>>          prompt="Password [$device ($luksname)]:"
>>          if [ ${#luksname} -gt 8 ]; then
>>              sluksname=${sluksname##luks-}
>
> Very nice! Thank you!
>

I think attached patch still makes sense in addition to (or probably
even instead of) this one.

[-- Attachment #2: dracut-plymouth-password-retries.patch --]
[-- Type: text/x-patch, Size: 977 bytes --]

From: Andrey Borzenkov <arvidjaar@gmail.com>
Subject: [PATCH] cryptoroot-ask: limit number of plymouth password requests

Set number ot retries to 5 to align with non-plymouth case.

Signed-off-by: Andrey Borzenkov <arvidjaar@gmail.com>

---
 modules.d/90crypt/cryptroot-ask.sh |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/modules.d/90crypt/cryptroot-ask.sh b/modules.d/90crypt/cryptroot-ask.sh
index 9c99ccc..4a65c23 100755
--- a/modules.d/90crypt/cryptroot-ask.sh
+++ b/modules.d/90crypt/cryptroot-ask.sh
@@ -92,7 +92,7 @@ else
         # flock against other interactive activities
         { flock -s 9; 
             /bin/plymouth ask-for-password \
-                --prompt "$prompt" \
+                --prompt "$prompt" --number-of-tries=5 \
                 --command="/sbin/cryptsetup luksOpen -T1 $device $luksname"
         } 9>/.console.lock
         
-- 
tg: (664193e..) upstream/passwd-retries (depends on: upstream/crypto-plymouth)

Re: [PATCH] cryptroot-ask: check that plymouth is running before use

[Harald Hoyer]

Am 11.03.2011 10:43, schrieb Andrey Borzenkov:
> I think attached patch still makes sense in addition to (or probably
> even instead of) this one.

also ok for me

Re: [PATCH] cryptroot-ask: check that plymouth is running before use

[Harald Hoyer]

Am 11.03.2011 10:43, schrieb Andrey Borzenkov:
> On Fri, Mar 11, 2011 at 11:41 AM, Harald Hoyer <harald-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> wrote:
>> Am 10.03.2011 23:54, schrieb Andrey Borzenkov:
>>> Otherwise there is no way to skip pasword prompt. --has-active-vt
>>> seems to correctly catch also the case when plymouthd is started
>>> but splash is disabled.
>>>
>>> Signed-off-by: Andrey Borzenkov <arvidjaar-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
>>>
>>> ---
>>>  modules.d/90crypt/cryptroot-ask.sh |    5 ++---
>>>  1 files changed, 2 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/modules.d/90crypt/cryptroot-ask.sh b/modules.d/90crypt/cryptroot-ask.sh
>>> index 5150688..9c99ccc 100755
>>> --- a/modules.d/90crypt/cryptroot-ask.sh
>>> +++ b/modules.d/90crypt/cryptroot-ask.sh
>>> @@ -80,9 +80,8 @@ if [ -n "$(getarg rd.luks.key)" ]; then
>>>      rmdir "$mntp"
>>>      unset mntp keypath keydev
>>>  else
>>> -    # Prompt for password with plymouth, if installed.
>>> -    # Should we check if plymouthd is running?
>>> -    if [ -x /bin/plymouth ]; then
>>> +    # Prompt for password with plymouth, if installed and running.
>>> +    if [ -x /bin/plymouth ] && /bin/plymouth --has-active-vt; then
>>>          prompt="Password [$device ($luksname)]:"
>>>          if [ ${#luksname} -gt 8 ]; then
>>>              sluksname=${sluksname##luks-}
>>
>> Very nice! Thank you!
>>
> 
> I think attached patch still makes sense in addition to (or probably
> even instead of) this one.

pushed both