the
sub-command. The case requests micro/patch binding.
I'm happy with the case as specified so it gets my +1.
--
Darren J Moffat
...?
While odd it is consistent with the rest of ZFS. You need SYS_CONFIG to
do disk or pool level stuff if you don't have a ZFS delegation.
--
Darren J Moffat
it clear that both files and directories, and all
other types of filesystem objects are supported here, and that it do so
by explicitly saying file and directory.
--
Darren J Moffat
On 29/03/2010 19:21, Don Cragun wrote:
On Mar 29, 2010, at 10:42 AM, Darren J Moffat wrote:
On 29/03/2010 18:30, Don Cragun wrote:
+ + Indicates the file/directory was added in the later dataset
+ - Indicates the file/directory was removed in the later dataset
Nexenta ship ?
Nexenta is mostly a Debian with OpenSolaris kernel (and some other
parts). The utilities userland is essentially the same as Debian Linux
so mostly GNU.
--
Darren J Moffat
.
--
Darren J Moffat
, with a default of false. This case proposes to
change the default to true.
I'm happy with this, gets my +1 as specified.
--
Darren J Moffat
, when ksh93 is the profile
shell ?
Why would I want to use ksh93 builtins if I have /usr/gnu/bin explicitly
in my path ? Are the ksh93 builtin versions 100% compatible in all
respects with the GNU ones ? If so then I wonder why we are even
shipping the GNU ones.
--
Darren J Moffat
On 18/03/2010 15:58, Jennifer Pioch wrote:
On Thu, Mar 18, 2010 at 4:41 PM, Darren J Moffat
darrenm at opensolaris.org wrote:
Maybe I don't understand enough about ksh93 (since I'm a zsh user for
interactive shell work) but I don't understand what this case is about.
What benefit does
, that is all I needed to know. I'm happy to see that
you and Glenn are working on it.
--
Darren J Moffat
providing builtins for that. Since we have no /usr/gnu/bin entries in
exec_attr anyway there is no issue with pfexec and this case.
--
Darren J Moffat
. As per [4]:
Nit, this case is actually asking for EOF and removal from OpenSolaris
not EOL.
Other than that I see no issues with this case and I support the removal
so it gets my +1.
--
Darren J Moffat
to use 32. It provides sufficient Linux compatibility and is
inline with standards recommendations.
--
Darren J Moffat
are ignored.
This is a very useful feature and I know of several customer deployments
where this would have been very valuable. Glad to see it.
--
Darren J Moffat
requirements.
I'm happy with that, particularly since there are a couple of existing
user_attr(4) keywords (idlecmd, idletime) that are probably EOF given
CDE is no longer shipped.
The case gets my +1 as specified.
--
Darren J Moffat
+ hits I got) - one of the main barriers to
changing that was NIS+ which has now been removed.
--
Darren J Moffat
User
profile and the use of gksu.
--
Darren J Moffat
sysconfig(2) for.
The above was on onnv_132 with $SHELL as /usr/bin/zsh, if ksh93 is used
then getconf is a shell built in but the same behaviour is seen because
it eventually calls out to /bin/getconf anyway.
--
Darren J Moffat
space.
--
Darren J Moffat
because there is no
integrated bluetooth stack on Solaris yet that could possibly connect to
it. BlueZ would just confuse users into thinking we had finally got a
bluetooth stack when we don't.
--
Darren J Moffat
the default initial cwnd.
This case does not change the default used in Solaris. An app
must still use the socket option to change the initial cwnd.
Where does 16 come from ? Why not 32 or 64 ?
--
Darren J Moffat
It is a little unfortunate that the attribute doesn't actually mean the
file is sparse, but given the primary motivation for this is SMB support
it seems reasonable so it gets my +1.
--
Darren J Moffat
.
Is there any reason why GUPnP can't be deliver for OpenSolaris ? There
are already other UPnP implementations delivered as part of
Coherence/Moovia.
--
Darren J Moffat
integrating this into Nevada
before the window of opportunity for S10U9 soak/backport closes.
Thanks,
Dan
On 01/28/10 10:31 AM, Darren J Moffat wrote:
Having discussed this offline with some people the following is my
suggestion to move this forward.
Samba can integrate with a private copy
in changing the system
libldap.
While this is less than ideal given the desire to quickly get an updated
Samba into Solaris 10 U9, where there is no in kernel CIFS server, I
think this is the best that this project team can be expected to do.
--
Darren J Moffat
).
--
Darren J Moffat
of any
applications that would depend on using chmod(2) while preserving an ACL
- since most apps don't know about ACLs and most apps (rather than code
system utils) don't call chmod(2).
So I'm happy with this and it gets my +1.
--
Darren J Moffat
?
--
Darren J Moffat
Jan Setje-Eilers wrote:
Darren J Moffat wrote:
Enrico Perla wrote:
Darren J Moffat wrote:
What is the method credential section of the SMF manifest used to
start the vbios service ? ie what user/group id does it run as and
what privileges(5) does it require.
It runs as root and needs
Jan Setje-Eilers wrote:
The case reflects the single daemon. In fact combining them was driven
by the inevitable reaction that multiple daemons would provoke.
Thanks that wasn't obvious to me initially. In that case my questions
have been answers to my satisfaction.
--
Darren J Moffat
I don't see any issues with this. The case answered everything I
expected, including changing the recommendation to deliver in /lib/.
So I'll happily give this my +1.
--
Darren J Moffat
+1 for the case as specified. I particularly like the change so that
successful logins are audited by default as soon as auditing is enabled.
--
Darren J Moffat
. The above information is great background but the project team
should pick one as their proposal (I think you are saying option 1) and
ask the ARC's opinion.
--
Darren J Moffat
. Timeout is 1/14/2010.
Even though this is a change in behaviour I don't see any problems with
this since it is an edge case feature that was introduced for the submitter.
So +1 from me.
--
Darren J Moffat
there is good motivation for an EOF and
removal. I understand the desire to remove any 10BASE-2 or 10BASE-T
support but there are cards out there driven by this driver that support
100BASE-T (3Com 3c905B 100BaseTX Cyclone, 0x9055).
--
Darren J Moffat
to
be USB attached now (even ones internal to some laptops).
--
Darren J Moffat
just don't see these parts as continuing to be
relevant enough to warrant any significant development effort spent on
their behalf.
I'm still not convinced that removing it is worth alienating people that
have such hardware given the (once) popularity of this device.
--
Darren J Moffat
hanging out of the laptop to a reader -
something that might be important when in the field uploading from
your cards to the laptop. Sure most people doing this are probably
using a Mac or Windows but at least some of us do use OpenSolaris for this.
--
Darren J Moffat
a full review in
this case since the issue isn't with the core architecture of Samba but
an issue with a dependency.
--
Darren J Moffat
While I think it is unfortunate we need yet another special uid/gid for
this it seems like the only workable solution (I'd already discussed
this offline with the project team). So I'm happy to given this case a
+1 as specified.
--
Darren J Moffat
.
The new privilege is called basic_test.
This makes sense as a debug functionality. Given this is for debug I
don't actually think an ARC case was required unless the intent is to
document this. So I'm happy to give this a +1.
--
Darren J Moffat
/ganglia.conf if you want
users to need to manually enable it). That will be cleaner and less
surprising for customers and safer for the package.
I agree this is a much better solution.
--
Darren J Moffat
I'm happy with the final spec.
--
Darren J Moffat
entered (assuming it was available) and password based
preauth would be attempted. If that complicates things too much or
there are other reasons to allow for two instances of pam_krb5 in the
stack then I'm happy with the case as specified.
--
Darren J Moffat
? Are files
created as this new user ?
Why are two SMF services required ? Does one depend on the other or do
both need to be enabled ?
Is there ever a need to run gmetad or gmond by hand as an administrator
? If not they should be in /usr/lib/ganglia/
--
Darren J Moffat
is I didn't see an updated smbadm(1M) man page - is
that because there are no visible interface changes to it ?
--
Darren J Moffat
or not
depends on wither or not it would cause the PKCS#11 token to record a
failed login attempt or if it would cause a Kerberos failed login attempt.
So I'd say be on the safe side and if pkinit is specified don't use
PAM_AUTHTOK at all for authenticating to the PKCS#11 token.
--
Darren J
it well!
I think the rationale is sound and I see no point in continuing to
support this hardware in future releases.
Assuming the release binding for this is minor (and not patch) then I'm
happy to give a +1 as specified.
--
Darren J Moffat
nice to see it doesn't require a pool or filesystem version
bump to use it (being familiar with this area of code that makes sense
to me).
--
Darren J Moffat
I'm happy with this case as specified so it gets my +1.
--
Darren J Moffat
anyway ?
--
Darren J Moffat
to implementation
details of the program.
--
Darren J Moffat
Garrett D'Amore wrote:
Darren J Moffat wrote:
This case looks reasonable as specified, however I do have one question.
Given hal is (or at least has been) used on Linux/BSD systems do they
already have this or a similar ioctl ? Is there any value in being
compatible or is this an area
I believe we are still waiting on a final spec for this case.
Specifically is the intent to add a 'pkinit' module option to the
existing pam_krb5 module or add a pam_krb5_pkinit module.
--
Darren J Moffat
Tavitayya Varanasi wrote:
Hi Darren,
Please look below for embedded answers.
Darren J Moffat wrote:
Why does this need a top level directory in /usr/ ? The only time we
should need those is when there is a hard requirement to be able to
support multiple incompatible versions
directory belongs but maybe following the
Cacacti case it should be /usr/share/ganglia/web.
What is the purpose of ganglia-monitor ?
This case as does not met the filesystem layout or SMF requirements for
Solaris and as such should not be approved as currently specified.
--
Darren J Moffat
uncommitted paths:
Maybe then the problem is that the default $PATH shouldn't have
/usr/gnu/bin in it.
I have no problem with removing these from the default install, but it
seems the GNU variants should still be available in a non default
package - but I can't seem to justify why.
--
Darren J
to be a fast-track.
--
Darren J Moffat
.
The real architectural problem though is that the desire to remove these
come from /usr/gnu/bin being higher in the default PATH. We should fix
the root cause of the problem.
--
Darren J Moffat
With the updated name I have only one possible small issue.
You said the driver is hardcoded to 512 byte blocks, yet the industry is
moving towards 4k blocks. Is that likely to be an issue for devices
driven by 'blkdev' ?
--
Darren J Moffat
Garrett D'Amore wrote:
Darren J Moffat wrote:
With the updated name I have only one possible small issue.
You said the driver is hardcoded to 512 byte blocks, yet the industry
is moving towards 4k blocks. Is that likely to be an issue for
devices driven by 'blkdev' ?
I don't think so
m_nblks;
uint32_t m_blksize;
boolean_tm_readonly;
};
I'm happy to give the case a +1 now.
--
Darren J Moffat
when the operating system stops without warning for example
on a panic or on power loss without a system UPS.
The DDRdrive X1 when used as a SLOG for ZFS may well require a UPS.
--
Darren J Moffat
don't really see that as arc review.
If this were a full case on Sun designed hardware and software I'd be
TCA strong verging on TCR to have the functionality redesigned to not
require the command for backup. However it is a third party device so
it is what it is.
--
Darren J Moffat
George Vasick wrote:
Darren J Moffat wrote:
So will all possible code that the GNU backend can build also be able
to be built with the Studio backend ?
all possible is a pretty big claim. The answer is a qualified yes. We
designed the product to be 100% compatible. There is always
there is. The project team has also
been asked (since they are Sun engineers) to check with Sun legal (this
isn't really an ARC requirement more a C-Team/P-Team one but the ARC is
asking the question now since they are first to review).
--
Darren J Moffat
to the GNU backend that would cause the Studio backend
to fail in error (rather than silently ignore) ?
Basically are they 100% compatible in there interfaces - if so then I
rescind my objection to it being the default.
--
Darren J Moffat
are required to setup a filter programatically ?
DJM-6 What RBAC rights profile will soconfig(1M) be in and what
privileges will it run with ?
--
Darren J Moffat
, defaulting to the Studio backend is a really bad idea.
Including the Studio backend is a great idea but the default should be
the default backend from upstream.
--
Darren J Moffat
the description of the tool and the fact it says attacks I feel
this is a very different case to including tools like nmap which are
scanners.
--
Darren J Moffat
but please don't add
new subdirectories under /usr/bin/ the only subdirs we should have
below /usr/bin are the $ISA specific ones. /usr/bin/X11 was a mistake
from ancient history that shouldn't be repeated.
--
Darren J Moffat
: 256 file descriptors
0: S_IFCHR mode:0620 dev:387,0 ino:1554593710 uid:23805 gid:7 rdev:24,1
O_RDWR|O_LARGEFILE
/dev/pts/1
offset:1469081
That looks great to me.
--
Darren J Moffat
separate modules will actually
turned out to be a problem. So I would prefer a pkinit module option,
that should be trivial to implement.
--
Darren J Moffat
Wyllys Ingersoll wrote:
Darren J Moffat wrote:
Wyllys Ingersoll wrote:
Gary Winiger wrote:
My personal recommendation: Develop a pam_pkinit (or similarly
named) module
with a separate man page. Have that man page describe the interactions
between pam_pkinit and pam_krb5.
Thanks
Jyri Virkki wrote:
Darren J Moffat wrote:
All system uid and gid that are used in packages included in the
distribution are allocated out of the reserved space (0-99), there is
lots of prior precedence for this.
[...]
During the package creation, there is file in which we mention whether
on the same line as mode: or on its own line.
If that is done then an offset of 0 should be printed for
things that are seekable.
--
Darren J Moffat
not be consuming
pfiles output, but possibly using procfs directly.
Indeed but that should imply we deliberately make the output worse
than it is just now, so putting offset on a new line by its own would
be my preference.
--
Darren J Moffat
be done on the system without a user building
wacom on there own ? It doesn't feel complete to me.
--
Darren J Moffat
Alan Coopersmith wrote:
Darren J Moffat wrote:
Alan Coopersmith wrote:
I am sponsoring this fast-track for Pengcheng Chen of the USB team.
The timer is set for next Thursday, November 19. The release binding
requested is Patch. A copy of the usbwcm(7m) draft man page is
available
.
--
Darren J Moffat
to design this for you, ask for advice in the
SMF community smf-discuss at opensolaris.org.
--
Darren J Moffat
it to run as root ?
--
Darren J Moffat
Spoorthy H S wrote:
Darren J Moffat wrote:
To monitor some system services, we start nrpe from root. Without
root privilage also, we can start this service.
That doesn't answer the question I asked. Why does nrpe need to run
as root ? If it doesn't need to then it should not make it so
Amit Gupta wrote:
On 11/11/09 22:34, Darren J Moffat wrote:
The filelayout looks fine.
What uid/gid and privileges does SMF start collectd with ? (ie what is
the method credential section in the SMF manifest).
SMF will start collectd as root.
Why does collectd need to run as root ?
What
amit.gupta at sun.com wrote:
Darren J Moffat wrote:
Amit Gupta wrote:
On 11/11/09 22:34, Darren J Moffat wrote:
The filelayout looks fine.
What uid/gid and privileges does SMF start collectd with ? (ie what
is the method credential section in the SMF manifest).
SMF will start collectd
package. If it is user package, in that case
we cannot copy files to directories directly under / other than /usr.
Not after SXCE is stopped being built you don't because IPS does not
make that distinction.
The new filesystem layout looks much better.
--
Darren J Moffat
is the method credentials used to start the service ? What
uid/gid does it run with, does it run with any privileges (other than
basic - and does it even need those ?)
--
Darren J Moffat
Why is SUNWipfh (which is a package name) listed as an imported interface ?
What is actually being imported from IPfilter ?
--
Darren J Moffat
The filelayout looks fine.
What uid/gid and privileges does SMF start collectd with ? (ie what is
the method credential section in the SMF manifest). Hopefully this need
to run as root with all privileges - ideally it should run as daemon
with basic privileges (or less).
--
Darren J Moffat
the SUNWipfh package really shouldn't be a package dependency since
it only has header files in it. If nrpe really uses the IPfilter header
files at runtime then I'm very worried.
--
Darren J Moffat
Amit Gupta wrote:
On 11/06/09 15:13, Darren J Moffat wrote:
Is collectd really that huge and important that it deserves its own
place under /usr ? Do Linux distros do this (I doubt it because the
FHS and LSB both very much frown on the new dirs in /usr).
Can this be:
/usr/lib
Mark A. Carlson wrote:
Has this case converged? It may be approved at today's LSARC if nobody
needs more time.
No it hasn't it needs a new spec with at least a filesystem layout that
matches what we do on Solaris.
--
Darren J Moffat
for a given user.
A generic change though of allowing the user to pick which auth stack
they want to run (ie a set of modules configured by an admin) is a
different mater though. The work on pam_eval and per user stacks would
be helpful in that though.
--
Darren J Moffat
I'm happy with the latest spec that has been proposed. I think this is
sufficient for now and it doesn't preclude adding module options or a
krb5.conf stanza (or even user_attr(4) name=value pairs) to control this
in the future.
--
Darren J Moffat
I'm submitting this as closed approved automatic, it has already been
reviewed by core ZFS and kernel scheduling teams. Given the taxonomy
of the new interfaces I believe this is sufficient review and closed
approved automatic is sufficient. If any member disagrees then I'll
set a fast-track
even though I think
delivering nagios is good idea.
--
Darren J Moffat
/checkresults
/usr/var ? Again this is wrong I suspect these should all be below
/var/nagios/
As specified this case violates many of the rules about filesystem
layout, so a big -1 from me on the filesystem layout even though I think
delivering nagios is good idea.
--
Darren J Moffat
be Project Private.
The interface is the SMF FMRI not the xml file. What is the FMRI ?
Does the FMRI for nrpe depend on the nagios FMRI on the local machine ?
What user/group/privileges does the SMF manifest start nrpe with ?
--
Darren J Moffat
is
maybe too much. So yes the existing check that ConsoleKit does will
suffice for now.
--
Darren J Moffat
.
--
Darren J Moffat
1 - 100 of 840 matches
Mail list logo
