The boot flip in compute.subiquity.tmpl addressed port 3002. xcatd's install
monitor listens on site.xcatiport, so a cluster that moves the port loses the
flip and every node PXE-loops back into the installer. The flip now reads
site.xcatiport and keeps 3002 as the default. TABLEBLANKOKAY, because the key is
optional and a plain TABLE lookup of an absent key fails the whole template.
The flip also counted any reply as an accepted request. It now requires the
monitor's "ready" greeting before it sends "next", and "done" afterwards, so a
different service on that port is not read as a flipped node.
subiquity_nfsroot_server in debian.pm called getipaddr without a family. A
dual-stack management node answers with its IPv6 address, and casper takes
everything after the first colon in nfsroot= as the path, so the live filesystem
never mounts. It now asks for IPv4, as dhcp.pm and mknb.pm do.
The DNS setup wrote the xcatmaster name as a nameserver when getent found no
address, which is the case the step exists to prevent. It now keeps the
resolv.conf DHCP gave the live installer.
ubuntu_subiquity_boot_flip.t, debian_subiquity_boot_params.t and
ubuntu_resolvconf_ip.t fail on the parent commit and pass here.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
EL10 defines no default crash kernel reservation for riscv64, so the
installer's kdump add-on falls back to writing the literal
"crashkernel=auto" into the boot loader arguments of the installed
system. EL10 kernels dropped support for that value: nothing is
reserved, but the string is on the command line, so kdump.service passes
its condition and then fails on every installed riscv64 node.
Turn the add-on off in the riscv64 templates. Nodes come up with kdump
inactive instead of failed, and a node that should take crash dumps gets
a real reservation the usual way, through linuximage.addkcmdline or
bootparams.addkcmdline.
The EL10 anaconda has no RISC-V EFI platform: on riscv64 it asks for the
x86 UEFI boot loader packages (grub2-efi-x64, shim-x64), which do not
exist, and registers the UEFI boot entry as \EFI\<distro>\shimx64.efi,
so a kickstart install stops at the missing packages and, once tolerated,
leaves a system the firmware cannot boot.
Add riscv64 templates for the rocky10/rhels10 compute and service
profiles that are the shared templates plus %packages --ignoremissing and
a %post fix-up, riscv64 package lists that add grub2-efi-riscv64 and
efibootmgr to the shared lists, and the fix-up itself
(post.rhels10.riscv64): it copies \EFI\<distro>\grubriscv64.efi to the
removable-media path \EFI\BOOT\BOOTRISCV64.EFI and re-points the UEFI
boot entry at grubriscv64.efi, so the installed node boots from disk with
or without usable NVRAM. Reinstalling a node replaces that entry instead
of adding another one.
The %post of a kickstart is a single shell script, into which xCAT
splices every #INCLUDE, and post.rhels10/post.rhels8 end it with
"exit 0"; the fix-up is therefore included ahead of them. The shared
templates and other architectures are unchanged.
Provide the compute and service profiles for rocky10 and rhels10 on
riscv64: netboot pkglist/exlist/postinstall files and the service
otherpkgs lists for netboot and install. They mirror the x86_64 profiles
(every package resolves in the Rocky 10 riscv64 BaseOS/AppStream/CRB
repositories and the minimal-environment group exists there), with the
kbd keymap exclude spelled correctly, the duplicate man exclude dropped,
and goconserver pulled from the rh10/riscv64 dependency repository.
Two ways the Subiquity install could fail without saying anything useful.
The boot flip read from the install monitor with no timeout. A monitor that accepts the
connection and then never answers -- which #7759 shows is a state it gets into -- blocked the read
forever, and with it the Subiquity late-command and the install. The retry loop could not help:
it never reached the retry. Both reads now take -t 10, so five attempts are bounded at roughly two
minutes and end in the failure that is already logged. The regression test stands up a listener
that accepts and holds the connection; removing the timeouts fails it.
mkinstall resolved the install server for nfsroot and fell back to the name when that failed:
"getipaddr($instserver) || $instserver". The name is the original defect -- klibc's nfsmount
cannot resolve one -- so the node panicked "can't parse IP address" at boot, on the node, with
nothing reported on the management node. The management node knows at template time, so it says
so there and skips the node, as the other unrecoverable per-node conditions in this routine do.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
The comment blamed mawk: updateflag.awk needs gawk's |& and Ubuntu's /usr/bin/awk is mawk. That
is wrong on this path -- compute.subiquity.tmpl has installed gawk in the target since Ubuntu
20.04 support was added, so updateflag.awk would have found it.
The flip was going unanswered because xcatd's install monitor, which serves port 3002, dies and
is never re-forked. That is fixed in #7759. The /dev/tcp exchange here is hardening: it retries
and reports, rather than reinstalling in silence.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
Three passes of the same reasoning had accumulated: in the code, in the POD of the routines the
previous commit extracted, and again in the test headers. Say each once, where the reader needs it.
The POD blocks were the worst of it -- extracting subiquity_kcmdline() and
defer_syncfiles_to_postboot() moved the essays out of the routines but did not shorten them. The
toram paragraph ran eleven lines for one kernel argument; the exact systemd-shutdown message and
the size of the 24.04 layers are colour, not reasoning.
The template's two comments and the apt-sources pair are trimmed the same way: the sources.list
explanation lives in Template.pm, and the test says which case it is checking and points there.
That the tests still pass after rewriting the template is the point of the previous commit -- a
test matching its text would have broken here.
47 comment and POD lines removed, no behaviour change.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
Treating an empty noderes.installnic as "mac" skipped noderes.primarynic: a node
that configures only primarynic was installed with a netplan matching on mac.mac
that never renamed the interface, and a primarynic holding a MAC address was
ignored. The order xCAT uses everywhere else is installnic -> primarynic ->
mac.mac, with either attribute allowed to hold an interface name or a MAC
address, and xCAT::NetworkUtils::gen_net_boot_params already implements it for
the netboot kernel parameters.
Resolve it in Perl and hand the template the answer:
- xCAT::Template::subiquity_install_netcfg() delegates the order to
gen_net_boot_params and returns the netplan set-name (empty when the device is
matched by MAC alone) and the address to match on, with mac.mac parsed by
xCAT::Utils::parseMacTabEntry so |-separated, !hostname-suffixed entries
resolve to THIS node rather than to whichever entry came first.
- #SUBIQUITYINSTALLNIC# / #SUBIQUITYINSTALLMAC# render those two values.
installnic and primarynic are read blank-okay -- a node that sets neither is
normal, and is exactly what "match on mac.mac" means -- while mac.mac stays a
required lookup, as it was when the template read it directly.
- compute.subiquity.tmpl consumes the resolved values. No part of the fallback,
and no mac.mac parsing, is left in shell: it only chooses whether to emit
set-name, based on whether a name was resolved.
ubuntu_subiquity_installnic.t now passes: the six resolution cases, multi-entry
mac.mac, and the netplan the template's own late-command writes for a resolved
name and for none. ubuntu_subiquity_template.t asserts the template no longer
reads installnic/primarynic or parses MACs itself.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
Boot the live installer correctly. Add boot=casper so casper actually processes
netboot=nfs instead of scanning local disks and panicking, resolve the install server
to a literal IP because casper mounts the live filesystem with klibc's nfsmount which
has no resolver, and add toram so casper copies the squashfs into RAM and unmounts the
NFS source. That last one is what lets the node reboot at all: with the NFS root still
mounted, systemd-shutdown waits forever on an lvm/pvscan wedged in uninterruptible I/O
on it and the node never power-cycles into the disk it just installed. casper has no
cmdline knob for NFS mount options -- it parses only nfsroot= and takes the whole value
as the path -- so toram is its supported way to avoid a network root.
Write the installer's resolv.conf nameserver as an IP. glibc's resolver discards a
hostname given on a nameserver line, so the xcatmaster name left the installer, and
the in-target apt that inherits the file, with no DNS at all.
Add the online archive through apt `sources:` on classic-sources releases, where
Subiquity renders the target sources.list from the install media alone and in-target
apt cannot find packages the ISO does not carry. Deb822 releases are excluded: their
primary mirror already lands in ubuntu.sources, so legacy .list files would configure
the same suites twice.
Flip the node to local-disk boot from the live installer over bash's /dev/tcp instead
of relying on updateflag.awk, which needs gawk's |& coprocess while Ubuntu's
/usr/bin/awk is mawk. The exchange is checked, and a failure is recorded in the
install log rather than silently PXE-looping into another install.
Defer syncfiles to the postbootscripts on the diskful install path, so it runs on the
booted node with sshd up rather than inside the in-target chroot where the MN cannot
reach it.
Make the test harness's provision wait overridable through WAIT_FOR_PROVISION; the
default stays short so a boot loop still fails fast.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
Resolve installnic with the non-fatal #TABLEBLANKOKAY# token so an absent
noderes.installnic renders blank instead of aborting xnba generation, and treat an
empty installnic exactly like "mac": match the interface by its boot MAC address and
do not rename it. That is the same default EL and SLES already get, since their
statefull templates never reference installnic at all.
Nodes that do set noderes.installnic are unaffected -- the set-name branch is
unchanged. The diskless/netboot path never referenced installnic and is untouched.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
The scan wrote each disk into a file named after the identifier it
reported, wwn, path or neither, and read back the groups of one such
file only. Two disks that reported different identifiers therefore
never competed on their driver group: a disk without a WWN was dropped
as soon as another disk reported one, and when the last disk scanned
reported a WWN the readback opened the WWN files alone. A direct
attached boot disk that reports no WWN thus lost to a RAID volume that
reports one, which is the case the driver groups exist to decide.
Write every disk into the file of its driver group and keep the
identifier as the sort key inside that group, ranked so that a WWN
sorts ahead of a path and a path ahead of no identifier. The driver
group now decides first for every disk, the identifier still decides
between disks of one group, and no disk is dropped from the scan.
The device filter accepted sd, hd, vd and nvme names, so the xvd names
that a Xen guest presents never entered the scan. On such a guest the
whole detection ran on an empty list and the script fell through to the
xvda fallback, which takes the first Xen disk without looking at any of
them.
Accept the xvd names in the filter. A Xen disk now goes through the
same classification, kernel search and driver sort as any other disk,
so a guest with more than one disk gets a chosen disk rather than the
first one. The fallback stays for the case where the scan still finds
nothing.
The nvme branch of the filter is anchored at the same time, so a name
only matches when it starts with nvme.
The RHEL 10 support added a second copy of the script. The copy carries
the Xen fallback and a failure log that the common script does not, and
the common script carries the Intel RSTe/VROC fallback that the copy
does not. A RHEL 10 node whose OS disk is a VROC volume therefore fell
back to /dev/sda, and every later fix to disk selection had to be
written twice to reach both.
Fold the Xen fallback and the failure log into the common script and
delete the copy. The RHEL 10 installer includes the common script, as
the other installers already do. The failure log runs only when the
including script defines msgutil_r, because the subiquity path does
not.
The driver sort put the ahci and ata_piix controllers in the same
choice group as the PMC MaxRAID and megaraid_sas RAID controllers. On a
server with both, the sort could select a RAID data volume as the OS
install disk.
Move the RAID controllers to the second choice group. The direct
attached controllers hold the likely boot volume, and a server with
only RAID volumes still selects them from the second group. The SAS
host adapters move to the third group and every other driver to a new
fourth group, so the relative order of the remaining drivers does not
change. The RHEL 10 installer includes its own copy of the script, so
both carry the change.
Recovered from the lenovobuild branch.
The SLE AutoYaST profile sized the root partition with <size>auto</size>, which
lets YaST pick a size and can leave the rest of the disk unused. RHEL
(part / --grow) and Ubuntu (root uses the remaining space) both give root the
whole disk; make SUSE consistent by using <size>max</size>. Swap stays auto.
Recovered from the unmerged lenovobuild branch (9a8679f4).
Co-authored-by: Jarrod Johnson <10814490+jjohnson42@users.noreply.github.com>
On a node whose OS disk is an Intel RSTe/VROC software RAID (/dev/md/Volume0_0
or /dev/md/Volume0), disk auto-detection finds nothing and getinstdisk falls
back to the hard-coded /dev/sda. Prefer the VROC volume over that default.
Only fires when no install disk was otherwise selected and the VROC device
actually exists, so it cannot mis-select over a real disk and has no effect on
non-VROC systems. The original commit hooked into a lenovobuild-specific M.2
detection block absent from master; adapted to master's fallback point.
Not lab-validated (no VROC hardware available).
Recovered from the unmerged lenovobuild branch (c6c70e5).
The generated /opt/xcat/xcatinstallpost ran the install postscripts first and
only disabled the xcatpostinit1 service afterwards. If one of those postscripts
rebooted the node (for example a firmware or kernel update), the service was
still enabled, so on the next boot the whole postscript set ran again from the
start. Disable the service *before* running the postscripts so a mid-postscript
reboot cannot re-trigger them.
The disable decision reads OSVER (base variant only), RUNBOOTSCRIPTS and
NODESTATUS, which the included xcatinstallpost script would otherwise set.
Moving the INCLUDE below the disable block leaves those variables unset, so read
them up front (the included script re-reads them) and source xcatlib.sh so
msgutil_r is available; the disable decision is then identical to before.
The original change (lenovobuild 9184cbe0) touched only the legacy post.xcat.
Modern EL uses the systemd variants: post.xcat.ng (RHEL/CentOS 8 and 9) and
post.xcat.rhels10 (RHEL 10). Apply the same reorder to all three so the fix
actually takes effect on current platforms.
Lab-validated with a full AlmaLinux 9.8 stateful install (post.xcat.ng path):
the generated xcatinstallpost disables the service before running the
postscripts, the postscripts run exactly once, the node reaches "booted", and
with the default NODESTATUS=1 the service is left enabled just as before.
Recovered from the unmerged lenovobuild branch (original 9184cbe0), adapted and
extended to the .ng and .rhels10 variants.
Co-authored-by: Jarrod Johnson <10814490+jjohnson42@users.noreply.github.com>
cons/kvm drives the KVM guest serial console through a shared tmux session on
the hypervisor (tmux, unlike screen, is present on every libvirt host we target
incl. ppc64le hosts that ship no screen). The KVM host package lists were never
updated for that switch, so an xCAT-provisioned KVM host got 'screen' (now
unused) -- or nothing -- where tmux is required, leaving rcons broken.
Ship tmux in the KVM host pkglists:
- install/rh, install/ubuntu: screen -> tmux / add tmux.
- netboot/{alma,centos,rocky}: were symlinks to the RHEL 5.4 list; make them
real lists with tmux (the rhel5/6/7 lists keep screen).
- install/suse: add a KVM host list (SUSE had none) carrying tmux.
screen stays where it is still used: the Xen console, and the EOL rhel5/6/7 and
centos6/7 lists.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
The diskful service-node otherpkgs pkglist exists for every EL version and
both arches EXCEPT rhels10 ppc64le: install/rh/ has
service.rhels10.x86_64.otherpkgs.pkglist (and install/alma/ symlinks it) but
no ppc64le counterpart, while rhels5/6/7/8/9 and the netboot/ tree carry both
arches. xcat-test's SN_setup_case points a ppc64le service node's osimage
otherpkglist at install/<distro>/service.<ver>.ppc64le.otherpkgs.pkglist; with
the file absent the list resolves empty, so the otherpkgs postscript installs
nothing -- xCATsn never lands, xcatd never starts on the service node, and
every hierarchical dispatch fails with "<sn>:3001 No route to host /
Connection refused", cascading the whole *_hierarchy test suite.
Add install/rh/service.rhels10.ppc64le.otherpkgs.pkglist (xCATsn + the rh10
ppc64le goconserver, mirroring rhels9) and the install/alma symlink to it,
matching the existing x86_64 layout.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
When a node's install NIC is selected by MAC (installnic=mac), the generated
netplan was invalid: it named the ethernet device the literal string 'mac' and
set its MAC match to the full pipe-delimited list of all node MACs. The result
had no valid interface, so neither the installer environment nor the installed
system came up on the network. Match the install NIC by interface-name glob in
the installer, and in the target use only the primary MAC and omit set-name
when the install NIC is selected by MAC.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
The generated Subiquity autoinstall config defined no identity section. Newer
Subiquity treats a config without an identity as incomplete and stops at an
interactive "Continue with autoinstall?" confirmation, so the install never
proceeds unattended (the prompt's advice to add 'autoinstall' to the kernel
command line is a red herring -- it was already present). Provide an identity
with a password sourced from the cluster passwd table so the install runs
non-interactively.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
Daniel Hilst
Vinícius Ferrão
Markus Hilger
Andrey Kiryanov
Markus Hilger
Mark Gurevich