From 1be510f9529cb082f802408b472a77d074b394c0 Mon Sep 17 00:00:00 2001 From: Nicolas Vigier Date: Sun, 14 Apr 2013 13:46:12 +0000 Subject: Add zarb MLs html archives --- zarb-ml/mageia-dev/2012-February/011883.html | 129 +++++++++++++++++++++++++++ 1 file changed, 129 insertions(+) create mode 100644 zarb-ml/mageia-dev/2012-February/011883.html (limited to 'zarb-ml/mageia-dev/2012-February/011883.html') diff --git a/zarb-ml/mageia-dev/2012-February/011883.html b/zarb-ml/mageia-dev/2012-February/011883.html new file mode 100644 index 000000000..998ab4959 --- /dev/null +++ b/zarb-ml/mageia-dev/2012-February/011883.html @@ -0,0 +1,129 @@ + + + + [Mageia-dev] Error message while trying checkout a package?with mgarepo + + + + + + + + + +

[Mageia-dev] Error message while trying checkout a package?with mgarepo

+ Michael Scherer + misc at zarb.org +
+ Sun Feb 12 12:42:19 CET 2012 +

+
+ +
Le dimanche 12 février 2012 à 11:30 +0100, Oliver Burger a écrit :
+> Am 12.02.2012 11:16, schrieb Michael Scherer:
+> > Are you sure that you used the complete
+> > It seems that the key that was uploaded was not in the good format. Can
+> > you make sure that you used the complete ssh key, ie the whole line, not
+> > just the binary encoded blurb in the middle. Without the beginning,
+> > ( ie ssh-rsa/ssh-dsa ), there is no way it would work, since openssh
+> > cannot guess the format and the algo to use.
+> >
+> > So try again by using the complete line :
+> > ssh-rsa XXXx....XXXX foo at example.org
+> >
+> > And if this solve the problem, could mentors please ensure that newer
+> > packagers actually really understand how ssh is supposed to work before
+> > directing them to sysadmin for diagnosing the same problem over and
+> > over?
+> Guilty as charged, but actually Dimitrios was a bit faster then me and I 
+> didn't have that much time yesterday to try and work this out.
+> 
+> > ( alternative version : code proper error reporting in identity )
+> > ( 2nd alternative version : add a foolproof mgarepo command to create
+> > and upload keys into our ldap )
+> If any of those two should be done, including it in identity would be 
+> the better way, I think.
+
+> We have people aside from packagers with svn commit rights, e.g. the 
+> i18n team commiters.
+
+Indeed. 
+
+But we can do both, and I think that adding a command to mgarepo would
+help to enforce a better security practice ( ie, few people know that
+using ssh-agent can cause issue if someone is root on the server that
+you connect too, since this person can reuse your key by hijacking the
+agent ).
+
+> As an idea: instead of that input field or as an alternative provide a 
+> possibility to upload the pubkey file?
+> This well make sure, no half keys are uploaded.
+
+In fact, we already make sure that no half key are uploaded ( hence the
+lack of key after upload ), we just do not signal it.
+
+And having a specific input would be slightly more complex ( but doable
+), and less generic ( and I think Buchan wanted to avoid that, but maybe
+I am wrong ).
+
+-- 
+Michael Scherer
+
+
+ + + + + + + + + + + + + + + + + + + + + + + + + + +
+

+ +
+More information about the Mageia-dev +mailing list
+ -- cgit v1.2.1