John,
Respondes to the 2 GUI-related items below:
On 04/ 4/12 02:27 PM, John Fischer wrote:
Drew,
I looked at AI and GUI. Mostly nits.
usr/src/cmd/auto-install/manifest/ai_manifest.xml
- how protective do we need to be with the chap
passwords? I am concerned that these will
start appearing in AI manifests in clear text.
usr/src/cmd/distro_const/checkpoints/pre_pkg_img_mod.py
- To be consistent please add a comment to 524 or remove the comment
from line 729 and 773.
524 self.modify_dhcpagent()
...
729 # modify /etc/default/dhcpagent
730 self.modify_dhcpagent()
731
...
773 # modify /etc/default/dhcpagent
774 self.modify_dhcpagent()
775
usr/src/cmd/gui-install/src/disk_discovery_screen.py
- should this comment read as error #1?
319 # error #2 - required fields must be entered
This was a merge error. I've now removed all the numbers
from the comments.
usr/src/cmd/gui-install/src/disk_screen.py
- should this string be changed to reference iSCSI disk instead
of disk?
628 LOGGER.debug("Including disk: %s", disk.ctd)
ok - changed the 2 lines here that log identical messages
to be:
LOGGER.debug("Including (iSCS) disk: %s", disk.ctd)
and
LOGGER.debug("Including (non-iSCSI) disk: %s", disk.ctd)
Thanks,
- Dermot
Thanks,
John
On 04/ 4/12 09:52 AM, Drew Fisher wrote:
Good morning!
Dermot and I would like to get a second-round code review for the
iSCSI project.
https://cr.opensolaris.org/action/browse/caiman/drewfish/iSCSI2/
Due to the large amount of time between code review cycles and the
large amount of code pushed for the UEFI project, an incremental
webrev is difficult to cobble together. If people really need one we
can try to get one created but since the majority of the first-round
comments were nits or small requests (except for Jesse's ...), I'm
not sure how much value there would be for an incremental.
Jesse's request (for those not interested in scrolling back in the
email history) was to make the Target LUN optional. If the LUN is
specified, only show the one LUN. If the LUN was omitted show all of
the LUNs on the Target.
These bits are current in a second round of QE testing but I'm not
expecting any large issues to come out of the test cycle.
I'd like to get the review by Wednesday, April 11th. I'd like to ask
the same people to look again at specific sections:
Karen: Could I get you to look specifically at the text-installer
changes?
Niall/John: Could I get you both to look specifically at the GUI
changes?
Jesse: Can you look at code in $SRC/lib/install_target as that's
where all the calls to iscsiadm and libima.so exist
If anybody needs screenshots or ISOs, let me know.
Thanks!
-Drew and Dermot
On 2/15/12 10:30 AM, Drew Fisher wrote:
Good morning!
Dermot and I would like to get a first-round code review for the
following PSARC/CRs:
PSARC/2012/023
<http://psarc.us.oracle.com/Archives/CaseLog/arc/PSARC/2012/023> Interactive
Installation to iSCSI
6974246 <http://monaco.us.oracle.com/detail.jsf?cr=6974246> Automated Install
should provide mechanism for setting iSCSI initiator-id-node
7004719 <http://monaco.us.oracle.com/detail.jsf?cr=7004719> Opensolaris
LiveCD installation should give a GUI for installing Solaris onto iSCSI LUN
7004720 <http://monaco.us.oracle.com/detail.jsf?cr=7004720> Opensolaris text
installer should give a screen for installing solaris onto iscsi lun
7114789 <http://monaco.us.oracle.com/detail.jsf?cr=7114789> unlabelled iSCSI
drives and setting an iSCSI boot-device require special handling on SPARC
7121245 <http://monaco.us.oracle.com/detail.jsf?cr=7121245> iscsi paths don't
translate to bootable OBP paths/strings
7132111 <http://monaco.us.oracle.com/detail.jsf?cr=7132111> sample ai
manifest file does not mention setting whole_disk in disk target section
7132457 <http://monaco.us.oracle.com/detail.jsf?cr=7132457> Race condition in
AI involving the target discovery and the check for the new OS Device Name
7145512 <http://monaco.us.oracle.com/detail.jsf?cr=7145512> pulling iSCSI
information from the DHCP server does not support iSCSI boot
https://cr.opensolaris.org/action/browse/caiman/drewfish/iSCSI/
Karen: Could I get you to look specifically at the text-installer
changes?
Niall/John: Could I get you both to look specifically at the GUI
changes?
Jesse: Can you look at code in $SRC/lib/install_target as that's
where all the calls to iscsiadm and libima.so exist
Outstanding issues:
- QE has not yet started a test cycle. I've been working with
Angela Li on this. I'm hoping to start a test cycle soon
- ai_manifest.4 is still being edited. I've emailed proposed
changes to Alta who will translate them into actual English
- The fix for 7132457 is slated to be tested by PIT soon for
confirmation.
ISO / USB access can be found at Hudson:
http://indiana-build.us.oracle.com/view/iSCSI/ Choose the job for
the installer you want to look at and you'll see links along the
job's page pointing to the latest ISO / USB image
Screenshots of the GUI and Text-Installer (for those not interested
in firing up an entire ISO) can be found here:
http://mox.us.oracle.com/code/screenshots/gui/
http://mox.us.oracle.com/code/screenshots/ti/
If people want to try out the ISOs, let me know and I can provide
target IP/LUN information to use.
Thanks!
-Drew and Dermot
_______________________________________________
caiman-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
_______________________________________________
caiman-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
_______________________________________________
caiman-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
_______________________________________________
caiman-discuss mailing list
[email protected]
http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
