tcons opens an rcons console per node in a tiled tmux session. On tmux 3.x
(shipped by all currently supported platforms) it no longer works:
* The session was named "tcons.$$". Modern tmux treats "." as the
session:window.pane target separator, so "tmux attach -t tcons.$$" is
parsed as session "tcons", pane "$$" and fails with "can't find
session"/"can't find pane" -- tcons cannot attach to the session it
just created. Rename it to "tcons_$$".
* The per-node "tmux select-pane", "tmux split" and "tmux select-layout"
commands had no target. When tcons is run from inside an existing tmux
session (the common case) they act on the caller's current session
instead of the detached tcons session, splitting the operator's own
window. Target every command explicitly at the tcons session.
* Move the pane-border-status option inside the loop so the console title
is set on each pane as it is created.
Validated on tmux next-3.4: the reworked script builds a 3-pane tiled
session with no errors, whereas the previous version's dotted session name
is unaddressable ("can't find pane").
Recovered from the unmerged lenovobuild branch (originals 5107b6ba1,
cefefa7d1, 6c72a2707, e040d24bb).
Co-authored-by: Jarrod Johnson <10814490+jjohnson42@users.noreply.github.com>
(cherry picked from commit dda7605028)
replaycons chose the confluent console log directory
(/var/log/confluent/consoles) whenever the confetty binary happened to be
present. On a host that has confluent installed but runs conserver as the
active console service, that directory is empty and replaycons could not find
the log. Gate on the site table's consoleservice attribute instead -- the
same setting rcons uses -- so replaycons reads /var/log/confluent/consoles
only when confluent is actually the configured console service.
Recovered from the unmerged lenovobuild branch (originals 3ce4d81b3, 57f6ddd38).
Co-authored-by: Jarrod Johnson <10814490+jjohnson42@users.noreply.github.com>
(cherry picked from commit 21da014ed7)
Modern x86-64 UEFI firmware (including OVMF and recent iPXE) reports DHCP
client architecture 00:07, but the xNBA network-boot conditionals only matched
00:09. A UEFI node would load xnba.efi, re-request with user-class xNBA and
arch 00:07, fail to match, and be handed xnba.efi again -- an endless boot
loop. Accept 00:07 alongside 00:09 in the xNBA UEFI branches of both the
network-level (addnet) and node-level (addnode) dhcpd statements.
Validated on an OVMF UEFI client: it progresses from xnba.efi to the .uefi
genesis config and boots, instead of looping on xnba.efi.
Recovered from the unmerged lenovobuild branch (5d697b5c, 86b0e12e).
Co-authored-by: Jarrod Johnson <10814490+jjohnson42@users.noreply.github.com>
(cherry picked from commit 60bd6e7cd9)
The DNS-managed-networks check used my_ip_facing($net->{net}), which only
returns local addresses on the interface directly facing that network. A
nameserver that is a local address reachable via a different (routed)
interface was therefore treated as external, and the network was wrongly
dropped from ddns management.
Use thishostisnot(), which recognizes any local address regardless of the
interface, and keep the existing <xcatmaster> and site-nameserver special
cases.
Recovered from the unmerged lenovobuild branch (original 3c343f83), adapted
to preserve those special cases: lab validation showed the original replaced
them wholesale, which would exclude any network whose nameserver is the
literal <xcatmaster> token (thishostisnot('<xcatmaster>') is true).
Co-authored-by: Jarrod Johnson <10814490+jjohnson42@users.noreply.github.com>
(cherry picked from commit a0dfc3343a)
For each secondary NIC in NICSTOBRINGUP, doxcat backgrounds a loop that polls
`ethtool $nic` until the link reports "detected: yes" before running dhclient.
A NIC that is never connected keeps that loop (and its `sleep 5`) running for
the life of the genesis environment, leaking background shells and needlessly
churning.
Add a second exit condition so each loop also stops once /tmp/netinitted
exists, and touch that file once the boot NIC has completed network init. A
link that never appears no longer waits indefinitely.
Recovered from the unmerged lenovobuild branch (original 9ece12e4).
Co-authored-by: Jarrod Johnson <10814490+jjohnson42@users.noreply.github.com>
(cherry picked from commit f24457cddf)
The genesis boot script sleeps 20 seconds unconditionally with the comment
"wait for NIC initialization", but that wait exists solely for System P LE
(ppc64) hardware discovery -- the surrounding section is explicitly ppc64
only. Every x86_64 and aarch64 node therefore pays a needless 20s delay on
each genesis boot.
Move the sleep inside the `[[ ${ARCH} =~ ppc64 ]]` block so only ppc64 waits.
Applied to the base script and both dracut_105 (el, ubuntu) variants.
Recovered from the unmerged lenovobuild branch (original d0916d9d).
Co-authored-by: Jarrod Johnson <10814490+jjohnson42@users.noreply.github.com>
(cherry picked from commit 32149f10f4)
ddns.pm's DNS-update send has a 3-try retry loop, but it only retried on
NOTAUTH and -- on the final attempt -- the `next` fell out of the loop without
reporting failure. makedns then printed "Completed updating DNS records." and
"DNS setup is completed" even when the update was never accepted.
Two problems this causes:
- SERVFAIL was never retried. named routinely replies SERVFAIL transiently
right after a zone (re)load, before the zone is ready to accept dynamic
updates; the update succeeds a second or two later. This reliably breaks real
flat-cluster provisioning on el9/el10, where the test case's makedns -n runs
right after named reloads the zone with the new compute node.
- A persistent NOTAUTH/SERVFAIL (retries exhausted) was silently swallowed and
reported as success, so a genuinely failed DNS update looked like it worked.
Extract the batch send into send_ddns_update(), which retries NOTAUTH and
SERVFAIL (pausing on SERVFAIL), returns 0 only on NOERROR, and otherwise emits
the error and returns 1. Both batch-send sites now `return 1 if
send_ddns_update(...)`, so a real failure aborts add_or_delete_records instead
of falsely reporting success.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
(cherry picked from commit 9dd025bf05)
go-xcat's install list still named yaboot-xcat, which was obsoleted in the
xcat-dep 2.18 rebuild and is no longer published. It also lists the cross-arch
ppc64 genesis packages, which the arch-split 2.18 dep repo no longer places in
the x86_64 view an x86 management node reads. Because EL dnf is strict by
default, any one of these unresolvable names aborts the entire `dnf install`
transaction, so `go-xcat install` fails outright on EL9/EL10 even though every
package the node actually needs is available.
Drop yaboot-xcat from the install list (kept in the uninstall list so existing
installs are still cleaned up) and pass `dnf --setopt=strict=0` so a package
missing from the enabled repository is warned and skipped instead of failing
the whole install. The existing post-install smoke test still catches a
genuinely broken install.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
(cherry picked from commit 1b5938969c)
2.18.1 was already published to the EL yum repos on 2026-07-10
(xCAT-2.18.1-snap202607100105, from master commit f6deff08b, now
tagged 2.18.1 on this branch). A released version number is
immutable, so the next 2.18.x release from this branch is 2.18.2.
On the ubuntu-24.04 runner, dpkg-buildpackage aborted every xcat-core
package with 'Unmet build dependencies: build-essential:native', so zero
debs were produced, reprepro created no dists/noble, and 'apt-get update'
against the local file: repo failed with 'dists/noble/.../Packages File
not found' -> install_xcat failed. Add build-essential to the workflow
dependency install so the local package repo is actually built.
(cherry picked from commit f9717ea8da)
The xcat-dep APT repo now publishes only the noble (24.04) codename at
.../apt/latest/xcat-dep; the bionic/devel dist no longer exists, so
install_xcat failed on apt-get update/install. Run the workflow on
ubuntu-24.04 and point the apt source at latest/xcat-dep noble main.
Drop the apt-key step: apt-key is removed on Ubuntu 24.04 and the legacy
apt.key is gone from the xcat-dep dir; the repo already installs via
allow-insecure-repositories/allow-unauthenticated.
(cherry picked from commit 83323504a6)
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>
(cherry picked from commit c1db278e1a)
Requiring xCAT-genesis-base >= 2:%{version} coupled the flat xcat-core to a
matching genesis-base on every marketing-version bump: a 2.18.1 core refused
to install against the deployed 2.18.0 genesis-base. genesis-base changes
rarely and lives in the separate per-EL xcat-dep repo, so floor the requirement
at the fixed 2.18.0 baseline instead. Core version bumps no longer force a
genesis-base re-publish.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
(cherry picked from commit f6deff08bc)
xcatprobe subcommands (xcatmn, discovery, osdeploy) and hierarchy.pm load
xCAT:: perl modules but only add $XCATROOT/probe/lib/perl to @INC, not the
main $XCATROOT/lib/perl where perl-xCAT ships them. The RPM builder stopped
embedding those helpers, so /opt/xcat/probe/lib/perl/xCAT/ was absent on
installed systems and the subcommands were reported "Unsupported sub command".
Stage GlobalDef.pm, NetworkUtils.pm and ServiceNodeUtils.pm into a temporary
xCAT-probe tree and build its source tarball once, before the parallel target
builds fork, so workers only read a complete archive. Add package-payload
regression tests. Adopts the approach from upstream PR #7581.
The remaining buildrpms.pl changes are deployment-compliance refactoring
(signed, self-contained core repo output).
Closes#7579
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
(cherry picked from commit 0c3c44f28b)
buildrpms.pl stages source tarballs into $HOME/rpmbuild/SOURCES but only runs
rpmdev-setuptree in its one-time env-setup path. On a host where that never ran
(or $HOME/rpmbuild was cleaned) source staging fails with "SOURCES/...: No such
file or directory", no srpms/rpms are produced, and the run still exits 0 -- so
the deployable repo silently ends up empty. Create the rpmbuild tree up front so
a build no longer depends on prior manual setup.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
(cherry picked from commit 552f87d4c2)
buildrpms.pl now writes xcat-core.repo, mklocalrepo.sh and buildinfo.txt into
dist/<target>/rpms (templates ported from buildcore.sh), so the built tree is
directly deployable to xcat.org and cluster-test.pl no longer needs to collect
and re-createrepo the dist output.
The full package set is built on every arch (x86_64 and ppc64le alike), so each
arch produces a complete, self-contained xcat-core repo -- no need to copy the
noarch packages from the x86_64 build into the ppc repo. --package now replaces
the default set (so --package xCAT-genesis-base builds only genesis-base for the
dep pipeline). xCATsn added to \@native_pkgs so its arch rpm is located correctly.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
(cherry picked from commit b0509b9616)
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>
(cherry picked from commit 9ea3de343b)
cons/kvm ran `ssh -t <vmhost> screen ... <pty>`, which requires screen on every
hypervisor. Bare libvirt hosts (notably ppc64le power servers) ship tmux but not
screen, so goconserver's console fork failed and no console was captured.
Use tmux (present on all our libvirt hosts) for the shared detach/reattach
session, wrapping `virsh console` keyed on the domain name (== the xCAT node name
for KVM) -- stable across guest reboots, unlike the raw serial pty. The session
status bar is disabled so the captured console log stays clean serial.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
(cherry picked from commit e331290414)
build_xmldesc only recognised cpumodel "ppc64" (big-endian) when deciding to set
<os type arch='ppc64' machine='pseries'>, but ppc64le hosts report cpumodel
"ppc64le" (virNodeGetInfo model). A guest on a ppc64le hypervisor was therefore
emitted as an x86-style domain -- no machine type and with pae/acpi/apic -- and
libvirt rejected it: "machine type 'pseries-*' does not support ACPI".
Recognise "ppc64le" alongside "ppc64" for the arch/machine block, and omit the
x86-only pae/acpi/apic features on both. Little- and big-endian pseries guests
both use libvirt <os> arch "ppc64".
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
(cherry picked from commit 37666b92cd)
genesis-base now ships in the per-EL xcat-dep repo while genesis-scripts ships
in the flat xcat-core. The exact "= 2:%{version}-%{release}" dependency forced
core and dep to be republished in lockstep on every core rebuild (a new snap in
core could not be installed until the exact-matching genesis-base snap was also
published to the client's dep channel -- which broke a production client).
Relax to ">= 2:%{version}": epoch 2, version >= 2.18.0, any release/snap. RPM
ignores the release when the dependency omits it, so any same-version (or newer)
genesis-base satisfies it. Routine core snaps no longer require a genesis-base
re-publish; lockstep now only matters on version bumps and the initial publish.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
(cherry picked from commit 23a5877480)
The single flat xcat-core repo serves el8/el9/el10 from one build, but two spec
decisions were resolved at BUILD time and so were wrong for the other ELs that
share the same repo.
1. DHCP backend. xCAT.spec and xCATsn.spec selected the dhcp provider with a
build-time %if (rhel >= 10 -> kea, else dhcpd), so an el10 build wrongly
required kea on el8/el9 (and an el8/el9 build wrongly required dhcpd on
el10). Replace it with an install-time RPM rich dependency:
Requires: (kea if (system-release >= 10) else /usr/sbin/dhcpd)
Requires: (kea-hooks if (system-release >= 10))
dnf now resolves it per client: kea on el10+ (which dropped ISC dhcp from the
distro), dhcpd on el8/el9. SLES has no "system-release" provide, so the
condition is false there and it falls back to /usr/sbin/dhcpd, preserving
prior behavior. system-release is versioned per release package
(el10=10.x, el9=9.x, el8=8.x).
2. openssl. Make the xCAT-server openssl Requires uniform across EL (non-SUSE)
instead of el10-only, so the produced package set does not depend on which EL
the build host happened to be.
Also drop xCAT-genesis-base from the default @PACKAGES set in buildrpms.pl: its
initramfs bundles the build-chroot kernel/glibc and is therefore OS- and
arch-dependent, so it cannot ship in the single flat core. It is now built per
target by the xcat-dep pipeline (mockbuild-all.pl, via
`buildrpms.pl --package xCAT-genesis-base`) and ships in xcat-dep/rh<N>. The
explicit `--package xCAT-genesis-base` build path is retained.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
(cherry picked from commit 529c821177)
Building the genesis netboot image on a usr-merged host (openSUSE Leap
15, and increasingly EL) leaves the extracted root with /bin, /sbin,
/lib and /lib64 as real directories holding the same files that already
live under /usr/*. rpm then sees each binary twice (e.g. fs/bin/bash and
fs/usr/bin/bash) and aborts the install of xCAT-genesis-base with dozens
of file conflicts, so the package cannot be installed. After unpacking
the dracut image, fold those top-level directories into /usr and replace
them with relative symlinks, matching a normal usr-merged layout. The
loop is a no-op when the image already ships them as symlinks, so EL
builds are unaffected.
Signed-off-by: Daniel Hilst <392820+dhilst@users.noreply.github.com>
(cherry picked from commit 0f100de5ac)
Vinícius Ferrão
xcat2-backport-automation[bot]
Vinícius Ferrão
Daniel Hilst
Vinícius Ferrão