I'm trying to write a script to retrieve the source for the latest version of an RPM from git.centos.org - basically `yumdownloader --source` but using git repos instead of yum repos. Unfortunately, reliably identifying the latest version is proving to be quite challenging.<br />
<br />
In each RPM repo on git.centos.org, a single branch is used for each major release (eg. "c8" for CentOS 8.*), and commits for all minor releases (eg. CentOS 8.1 and 8.2) are mixed together in the same branch. Because of this, and the fact that multiple versions of the same RPM may be maintained for different minor releases simultaneously/asynchronously, the latest commit in a branch does not reliably correspond with the latest version of the RPM.<br />
For example, in <a href="https://git.centos.org/rpms/java-11-openjdk/tree/c8">https://git.centos.org/rpms/java-11-openjdk/tree/c8</a> :<br />
<a href="https://git.centos.org/rpms/java-11-openjdk/c/0b2558c84aff4c2c2fe0a0d3981b09ba93f312c5?branch=c8">https://git.centos.org/rpms/java-11-openjdk/c/0b2558c84aff4c2c2fe0a0d3981b09ba93f312c5?branch=c8</a> imported java-11-openjdk-11.0.7.10-1.el8_1<br />
before <a href="https://git.centos.org/rpms/java-11-openjdk/c/aa59cc3aacb503e3723eefd125b99c6c7681a253?branch=c8">https://git.centos.org/rpms/java-11-openjdk/c/aa59cc3aacb503e3723eefd125b99c6c7681a253?branch=c8</a> imported java-11-openjdk-11.0.6.10-4.el8<br />
and then no further versions were imported for a few months after that.<br />
<br />
To handle this, I currently parse the version number out of each "import" commit message, then use rpmdev-vercmp to compare the version numbers and identify the latest version number. However, since the version numbers in the "import" commit messages (and associated tags) do not include the Epoch, this falls apart for any RPMs that have changed their Epoch. There doesn't appear to be any way to rectify that problem short of extracting and parsing the spec file from every "import" commit.<br />
<br />
Could the scripts that generate these "import" commit messages and associated tags be updated to include the Epoch in the version number to fix this problem?
↧
0017831: Missing Epoch on "import" commits on git.centos.org makes it extremely difficult to reliably find the latest version
↧
0017832: kernel 3.10.0-1127.18.2.el7.x86_64 going to Grub mode after the yum update
we are in Kernel 3.10.0-1127.18.2.el7.x86_64 and after performing yum update and reboot it going to grub mode.<br />
<br />
yum update is installing 3.10.0-1127.19.1.el7.x86_64. <br />
<br />
We are observing this pattern only when we are updating from Kernel 3.10.0-1127.18.2.el7.x86_64 to 3.10.0-1127.19.1.el7.x86_64. <br />
<br />
Please help us to resolve this issue.
↧
↧
0017482: kworker/0:1+events using 33% CPU usage on a single core
After upgrading to 8.2.2004 of CentOS 8, I have noted there is now a single kworker process that is continuously utilising around 33% of a single CPU core:<br />
<br />
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND<br />
52 root 20 0 0 0 0 R 32.9 0.0 17:35.57 kworker/0:1+events<br />
<br />
Googling has revealed this as a known problem for a couple of years now, and from what I can see, users have identified the problem as being excessive ACPI GPE's being generated, which you can view from running a command such as:<br />
<br />
grep . -r /sys/firmware/acpi/interrupts<br />
<br />
However, when running the above, all the counters on each GPE entry are 0 on my system, which suggests that there is actually no ACPI GPE's being generated at all. Further more, running "iotop -b -o -t | grep -i kworker" also shows no unusual activity - in fact, the exact kworker process in question ("0:1" in this case) doesn't even show up on the iotop output.<br />
<br />
I have also added "acpi_enforce_resources=lax" to the Kernel boot params in an attempt to resolve, but not affect. I added this because I noticed that the "f71882fg" temp module was no longer loading after the upgrade, failing with this error:<br />
<br />
lm_sensors-modprobe-wrapper[2805]: modprobe: ERROR: could not insert 'f71882fg': Device or resource busy<br />
<br />
and adding "acpi_enforce_resources=lax" has now resolved this.<br />
<br />
If any further diagnostics are required, please let me know required commands to run and I will update with results when I can.<br />
<br />
Thank you.
↧
0017051: osinfo-db missing centos8
The current package osinfo-db-20190319-2.el7.noarch is missing centos8 from the database, even though it presently contains rhel8.<br />
<br />
$ osinfo-query os |grep rhel<br />
# OUTPUT TRIMMED<br />
rhel7.6 | Red Hat Enterprise Linux 7.6 | 7.6 | <a href="http://redhat.com/rhel/7.6">http://redhat.com/rhel/7.6</a> <br />
rhel7.7 | Red Hat Enterprise Linux 7.7 | 7.7 | <a href="http://redhat.com/rhel/7.7">http://redhat.com/rhel/7.7</a> <br />
rhel8-unknown | Red Hat Enterprise Linux 8 Unknown | 8-unknown | <a href="http://redhat.com/rhel/8-unknown">http://redhat.com/rhel/8-unknown</a> <br />
rhel8.0 | Red Hat Enterprise Linux 8.0 | 8.0 | <a href="http://redhat.com/rhel/8.0">http://redhat.com/rhel/8.0</a> <br />
<br />
$ osinfo-query os |grep centos<br />
# OUTPUT TRIMMED<br />
centos6.8 | CentOS 6.8 | 6.8 | <a href="http://centos.org/centos/6.8">http://centos.org/centos/6.8</a> <br />
centos6.9 | CentOS 6.9 | 6.9 | <a href="http://centos.org/centos/6.9">http://centos.org/centos/6.9</a> <br />
centos7.0 | CentOS 7.0 | 7.0 | <a href="http://centos.org/centos/7.0">http://centos.org/centos/7.0</a> <br />
<br />
This will make attempting to install a CentOS8 VM on a CentOS7 virt host fail when using the virt-install command with the os-variant switch.
↧
0017752: IDM not installable in 8-stream - No available modular metadata for modular package
<a href="mailto:root@ipa1">root@ipa1</a> ~ # dnf module install idm:DL1/{client,dns,server}<br />
Last metadata expiration check: 2:52:55 ago on Fri 25 Sep 2020 05:44:45 AM CEST.<br />
Dependencies resolved.<br />
====================================================================================================================<br />
Package Arch Version Repository Size<br />
====================================================================================================================<br />
Installing group/module packages: <br />
ipa-client x86_64 4.8.7-10.module_el8.3.0+482+9e103aab appstream 271 k<br />
ipa-healthcheck noarch 0.4-6.module_el8.3.0+482+9e103aab appstream 85 k<br />
ipa-healthcheck-core noarch 0.4-6.module_el8.3.0+482+9e103aab appstream 49 k<br />
ipa-server x86_64 4.8.7-10.module_el8.3.0+482+9e103aab appstream 522 k<br />
ipa-server-dns noarch 4.8.7-10.module_el8.3.0+482+9e103aab appstream 186 k<br />
…<br />
Is this ok [y/N]: y <br />
Downloading Packages: <br />
[SKIPPED] 389-ds-base-1.4.3.8-5.module_el8.3.0+473+53682548.x86_64.rpm: Already downloaded<br />
[SKIPPED] 389-ds-base-libs-1.4.3.8-5.module_el8.3.0+473+53682548.x86_64.rpm: Already downloaded <br />
…<br />
Running transaction check<br />
No available modular metadata for modular package '389-ds-base-1.4.3.8-5.module_el8.3.0+473+53682548.x86_64', it cannot be installed on the system<br />
No available modular metadata for modular package '389-ds-base-libs-1.4.3.8-5.module_el8.3.0+473+53682548.x86_64', it cannot be installed on the system<br />
No available modular metadata for modular package 'python3-lib389-1.4.3.8-5.module_el8.3.0+473+53682548.noarch', it cannot be installed on the system <br />
The downloaded packages were saved in cache until the next successful transaction.<br />
You can remove cached packages by executing 'dnf clean packages'.<br />
Error: No available modular metadata for modular package
↧
↧
0017833: Package Qt5Qml not available
I want to build AusweiApp2 from <a href="https://github.com/governikus/AusweisApp2">https://github.com/governikus/AusweisApp2</a> but cannot solve a dependency to Qt5Qml.<br />
Either I am overlooking a package or the package is missing Centos 8.
↧
0016717: Add support CentOS 8 in CPE oval and dict
Hi everybody<br />
<br />
Currently, there is no CentOS 8 platform in CPE when using <br />
<br />
# oscap -V<br />
OpenSCAP command line tool (oscap) 1.3.0<br />
...<br />
Community Enterprise Operating System 5 - cpe:/o:centos:centos:5<br />
Community Enterprise Operating System 6 - cpe:/o:centos:centos:6<br />
Community Enterprise Operating System 7 - cpe:/o:centos:centos:7<br />
--> there is no CentOS 8!<br />
...<br />
<br />
CPE name of CentOS 8 is cpe:/o:centos:centos:8 <br />
<br />
# cat /etc/os-release<br />
...<br />
CPE_NAME="cpe:/o:centos:centos:8"<br />
...<br />
<br />
So we need to adding CentOS 8 platform so that oscap can detect it!
↧
0017838: cron job for rhsmd gets terminated because settings moved away
cron.daily job for rhsmd gets terminated because settings are moved away<br />
<br />
this line is not working in /etc/cron.daily/rhsmd (it sets empty value for config variable)<br />
config=$(grep -E "^processTimeout" /etc/rhsm/rhsm.conf | grep -Po "[0-9]+")<br />
<br />
but it is working with different command:<br />
# subscription-manager config | grep --ignore-case "processTimeout"<br />
processtimeout = [300]<br />
<br />
every day cron.dai prints out to stdout (and sends email) with this info:<br />
### start<br />
/etc/cron.daily/rhsmd:<br />
<br />
sleep: missing operand<br />
Try 'sleep --help' for more information.<br />
/etc/cron.daily/rhsmd: line 19: 21531 Terminated /usr/libexec/rhsmd -s<br />
### end
↧
0016892: 8-stream: podman-manpages-1.4.2-5.module_el8.1.0+237+63e26edc.noarch conflicts with file from package docker-ce-cli-1:19.03.5-3.
Installation of podman is not possible on 8-stream if docker-ce is already installed because podman-manpages conflicts with docker-ce-cli for some really weird reason.<br />
<br />
This is a clear regression because on 8.0 they could co-exist peacefully.<br />
<br />
That issue is important as will clearly affect people trying to migrate from docker to podman, or those that are forced to use both due to incompatibiltiies.
↧
↧
0017839: need a git.centos.org repo for luarocks
needed for new Ceph work. Will be part of Ceph in the Storage SIG<br />
<br />
Thanks.
↧
0017841: [abrt] bash: kill_pid(): bash killed by SIGSEGV
Description of problem:<br />
Deliberate crash - bash -c 'kill -SEGV $$'<br />
<br />
Version-Release number of selected component:<br />
bash-4.2.46-34.el7<br />
<br />
Truncated backtrace:<br />
Thread no. 1 (8 frames)<br />
#1 kill_pid at jobs.c:3035<br />
#2 kill_builtin at ./kill.def:228<br />
#3 execute_builtin at execute_cmd.c:4167<br />
#4 execute_builtin_or_function at execute_cmd.c:4592<br />
#5 execute_simple_command at execute_cmd.c:4002<br />
#6 execute_command_internal at execute_cmd.c:765<br />
#7 parse_and_execute at evalstring.c:340<br />
#8 run_one_command at shell.c:1325
↧
0017840: Update for libssh-0.9.0-4.el8.x86_64 as it has bug. Refer libssh website https://www.libssh.org/2020/09/10/libssh-0-9-5/ .
Hello Team,<br />
<br />
We have CentOS Linux release 8.2.2004 (Core) with 4.18.0-193.el8.x86_64 version. We have monitoring tool installed on it and this monitoring tool needs to connect to client machine to fetch server details over the ssh. However due to bug in libssh-0.9.0-4 , we are seeing intermittent issue while connecting to client machine.<br />
I can see libssh community has release the updated version which is libssh 0.9.5 however I am unable to see this package on Centos 8.2 Base repo.<br />
<br />
I would like to check with you if there is any timeline for releasing this package on Centos 8 base repo.<br />
It caused us to move forward with Centos 8 version in prod environment.<br />
<br />
Pleases let me know if need more information.<br />
<br />
-Dushyant
↧
0015833: ipv6 router advertisements are disabled due to IPV6_AUTOCONF=no
I tested this with an Amazon VPC with working IPv6 configuration (the amazon-ami's work fine) using cloud-init 18.2_1.<br />
<br />
If one assigns an IPv6 address to the machine via DHCP6. Cloudinit sets up a configuration that apparently disables router advertisements due to side-effect. As a result the box has no default gateway and is unable to send IPv6 traffic (Network unreachable errors).<br />
<br />
I found someone else mentioning this issue here and their work-around: <a href="https://access.redhat.com/solutions/347693#comment-520173">https://access.redhat.com/solutions/347693#comment-520173</a><br />
<br />
And it basically came down to having IPV6_AUTOCONF=no which sets the sysctl setting, net.ipv6.conf.eth0.accept_ra=0. Disabling router advertisements.<br />
<br />
Overriding IPV6_AUTOCONF in /etc/sysconfig/network-scripts to yes enables the seemingly correct functionality. This appears to be set by cloud-init in /usr/lib/python2.7/site-packages/cloudinit/net/sysconfig.py.
↧
↧
0017843: Unable to install CentOS 8 in VMWare WS 16
Excuse me, but do you even test your software before public release? Typical first level tests would be install on baremetal as well as in Virtual Machines. We already fail here and maybe there are further problems.<br />
<br />
Anyway here are all details:<br />
System environment:<br />
Windows 10 build 1909 as host<br />
VMware 16.0.0 installed on Windows host<br />
CentOS 8: CentOS-8.2.2004-x86_64-minimal.iso<br />
<br />
First minor issue: <screenshot attached>. Testers should know what's important and bundle everything that's needed for the minimal installation instead of throwing "Problems in request".<br />
<br />
Second problem after I click ignore. This happens much later where the whole installer crashes. From VMware, I am not able to copy paste the whole error log but attached some screenshots with titbits of info. Of course, you can get the whole error log yourself, but just imply trying to install your buggy software yourself. I am 100% you'll be frustrated too why you can't install CentOS and scratch your head and ask "how did our team release this??".<br />
<br />
I expected CentOS to be better. Please fix the buggy software! <br />
<br />
Thanks
↧
0017847: krb5-libs version is not up to date with RHEL 8
We use krb5-libs as a dependency when compiling OpenLDAP for the LDAP Tool Box project.<br />
<br />
Current version of krb5-libs is 1.17 on CentOS 8 and 1.18 on CentOS 8, so our packages can not be installed on RHEL 8<br />
<br />
Could you update krb5 in CentOS 8 so packages built on CentOS 8 can work on RHEL 8 ?
↧
0017848: base repomd.xml signature outdated
On a server which has had CentOS 7.8 + CR packages installed, running "yum update" after 7.9 was released yields this output:<br />
<br />
# yum update<br />
Loaded plugins: fastestmirror, priorities<br />
Loading mirror speeds from cached hostfile<br />
epel/x86_64/metalink | 21 kB 00:00:00<br />
* base: ftp.funet.fi<br />
* epel: www.nic.funet.fi<br />
* extras: mirror.scaleuptech.com<br />
* updates: ftp.funet.fi<br />
base/7/x86_64/signature | 811 B 00:00:00<br />
Retrieving key from <a href="file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7">file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7</a><br />
Importing GPG key 0xF4A80EB5:<br />
Userid : "CentOS-7 Key (CentOS 7 Official Signing Key) <<a href="mailto:security@centos.org">security@centos.org</a>>"<br />
Fingerprint: 6341 ab27 53d7 8a78 a7c2 7bb1 24c6 a8a7 f4a8 0eb5<br />
Package : centos-release-7-8.2003.0.el7.centos.x86_64 (<span class="mention"><a class="user" href="https://bugs.centos.org/view_user_page.php?id=6483">@b</a></span>)<br />
From : /etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7<br />
Is this ok [y/N]: y<br />
base/7/x86_64/signature | 3.6 kB 00:00:00 !!!<br />
<a href="http://ftp.funet.fi/pub/mirrors/centos.org/7.9.2009/os/x86_64/repodata/repomd.xml:">http://ftp.funet.fi/pub/mirrors/centos.org/7.9.2009/os/x86_64/repodata/repomd.xml:</a> [Errno -1] repomd.xml signature could not be verified for base<br />
Trying other mirror.<br />
base/7/x86_64/signature | 811 B 00:00:00<br />
Retrieving key from <a href="file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7">file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7</a><br />
Importing GPG key 0xF4A80EB5:<br />
Userid : "CentOS-7 Key (CentOS 7 Official Signing Key) <<a href="mailto:security@centos.org">security@centos.org</a>>"<br />
Fingerprint: 6341 ab27 53d7 8a78 a7c2 7bb1 24c6 a8a7 f4a8 0eb5<br />
Package : centos-release-7-8.2003.0.el7.centos.x86_64 (<span class="mention"><a class="user" href="https://bugs.centos.org/view_user_page.php?id=6483">@b</a></span>)<br />
From : /etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-7<br />
Is this ok [y/N]: y<br />
<br />
.. and it will eventually output:<br />
Gpg Keys not imported, cannot verify repomd.xml for repo base<br />
<br />
Looking at <a href="http://mirror.centos.org/centos/7.9.2009/os/x86_64/repodata/">http://mirror.centos.org/centos/7.9.2009/os/x86_64/repodata/</a> I see:<br />
<br />
[TXT] repomd.xml 2020-10-29 20:03 3.6K <br />
[TXT] repomd.xml.asc 2020-10-26 16:51 811 <br />
<br />
meaning that repomd.xml was updated at some point, but the corresponding .asc signature wasn't.
↧
0017849: fasttrack repo does not exist
<a href="http://mirror.centos.org/centos/7.8.2003/">http://mirror.centos.org/centos/7.8.2003/</a> shows there's a fasttrack repository, but it does not exist in <a href="http://mirror.centos.org/centos/7.9.2009/">http://mirror.centos.org/centos/7.9.2009/</a><br />
<br />
I would suggest creating at least an empty fasttrack repo so that users' configuration would not break. Note that the fasttrack repo is referenced in centos-release-7-9.2009.0.el7.centos.x86_64.rpm's CentOS-fasttrack.repo, so presumably the repository should still exist.
↧
↧
0017846: Perl packages are missing in 7.9.2009's repo
Error when trying to install perl packages occurred today:<br />
<br />
yum install perl<br />
<br />
Error downloading packages:<br />
4:perl-5.16.3-295.el7.x86_64: [Errno 256] No more mirrors to try.<br />
4:perl-macros-5.16.3-295.el7.x86_64: [Errno 256] No more mirrors to try.<br />
4:perl-libs-5.16.3-295.el7.x86_64: [Errno 256] No more mirrors to try.<br />
1:perl-Pod-Escapes-1.04-295.el7.noarch: [Errno 256] No more mirrors to try.
↧
0017850: http://vault.centos.org/centos/7/updates/Source/repodata/repomd.xml vault returns 404
Vault has gone unavailable<br />
<br />
<a href="http://vault.centos.org/centos/7/updates/Source/repodata/repomd.xml:">http://vault.centos.org/centos/7/updates/Source/repodata/repomd.xml:</a> [Errno 14] HTTP Error 404 - Not Found
↧
0017844: SED OPAL kernel support is needed or disks don't get a good night's rest.
I've already reported this issue to Red Hat, since the problem is inherited through<br />
their kernel configuration. But am reporting it here in the hope that CentOS<br />
will be faster to fix it. That said, the rambling description goes something like:<br />
<br />
My hard drive is tired, but the kernel won't let it sleep. That's because<br />
my hard drive is one of those self encrypting drives. Which means it goes to<br />
sleep like every other hard drive. But when it wakes up, it likes to play the<br />
morning after game. It wants to know if its partner, the kernel, remembers<br />
its name. And if it doesn't, then Miss Disk will play hard to get. I don't<br />
know what happens next. But I do know its a total system crash. And that's bad.<br />
<br />
The good news is they don't need couples therapy. All Miss Disk needs is for<br />
the kernel to be compiled with CONFIG_BLK_SED_OPAL="y" ... I verified this the<br />
hard way. By recompiling a kernel myself. It was painful.<br />
<br />
If there is any bad news, it's that support for self encrypting OPAL drives<br />
can't be compiled as a standalone module. So I'm hoping the you can enable this<br />
flag for me.<br />
<br />
I'm not quite sure why this kernel flag is unset. I checked every Fedora release from<br />
Core 25 through Core 33 and all of them had this flag enabled already.<br />
<br />
<br />
### The problem is that without kernel support for SED OPAL drives.<br />
[<a href="mailto:ladar@crossroads">ladar@crossroads</a> ~]$ sudo /usr/local/sbin/sedutil-cli -s --prepareForS3Sleep 0 Admin1 /dev/nvme0n1<br />
<br />
<br />
Please enter password *********************************<br />
Error saving the password to the kernel errno = 25<br />
[<a href="mailto:ladar@crossroads">ladar@crossroads</a> ~]$<br />
###<br />
<br />
I get the above error, and if the system does go to sleep, it will slumber just fine, only to blink and<br />
crash when it wakes back up.<br />
<br />
### With kernel support for SED OPAL drives enabled.<br />
[<a href="mailto:ladar@crossroads">ladar@crossroads</a> ~]$ sudo /usr/local/sbin/sedutil-cli -s --prepareForS3Sleep 0 Admin1 /dev/nvme0n1<br />
<br />
<br />
Please enter password *********************************<br />
[<a href="mailto:ladar@crossroads">ladar@crossroads</a> ~]$<br />
###<br />
<br />
After this, I can put the system to sleep, and wake it back up again, all<br />
without issue.<br />
<br />
For those who stumble upon this bug report, once the kernel has OPAL support,<br />
the drive can be prepared for its nap automatically by a systemd service:<br />
<br />
### /etc/systemd/system/sedutil-sleep.service<br />
[Unit]<br />
Description=sedutil-sleep<br />
<br />
[Service]<br />
Type=oneshot<br />
ExecStart=/usr/local/sbin/sedutil-cli -n -x --prepareForS3Sleep 1 User1 <HASHED_PASSWORD> /dev/nvme0n1<br />
RemainAfterExit=true<br />
<br />
[Install]<br />
WantedBy=multi-user.target<br />
###<br />
<br />
In this example, the service file would have the hashed user password at<br />
locking range 1, and would supply that hash to kernel so it can properly<br />
wake up the drive when needed.<br />
<br />
### The CentOS and CentOS kernels I checked.<br />
[<a href="mailto:ladar@centos8">ladar@centos8</a> ~]$ cat /boot/config-4.18.0-193.28.1.el8_2.x86_64 | grep OPAL<br />
# CONFIG_BLK_SED_OPAL is not set<br />
[<a href="mailto:ladar@centos8">ladar@centos8</a> ~]$ cat /boot/config-4.18.0-193.19.1.el8_2.centos.plus.x86_64 | grep OPAL<br />
# CONFIG_BLK_SED_OPAL is not set<br />
[<a href="mailto:ladar@centos8">ladar@centos8</a> ~]$ cat /etc/os-release<br />
NAME="CentOS Linux"<br />
VERSION="8 (Core)"<br />
ID="centos"<br />
ID_LIKE="rhel fedora"<br />
VERSION_ID="8"<br />
PLATFORM_ID="platform:el8"<br />
PRETTY_NAME="CentOS Linux 8 (Core)"<br />
ANSI_COLOR="0;31"<br />
CPE_NAME="cpe:/o:centos:centos:8"<br />
HOME_URL="<a href="https://www.centos.org/"">https://www.centos.org/"</a><br />
BUG_REPORT_URL="<a href="https://bugs.centos.org/"">https://bugs.centos.org/"</a><br />
<br />
CENTOS_MANTISBT_PROJECT="CentOS-8"<br />
CENTOS_MANTISBT_PROJECT_VERSION="8"<br />
REDHAT_SUPPORT_PRODUCT="centos"<br />
REDHAT_SUPPORT_PRODUCT_VERSION="8"<br />
###<br />
<br />
When I recompiled a kernel for testing, I simply patched the kernel-plus.spec<br />
file. But that may not be ideal. I'm only providing a recompile/patch in case others<br />
want to try out this fix.<br />
<br />
On a separate, but related note of some importance, I couldn't find the <br />
kernel-plus-4.18.0-193.19.1.el8_2.centos.plus.src.rpm source RPM in any<br />
of the usual places. I had to dig just to find the <br />
kernel-plus-4.18.0-193.14.2.el8_2.centos.plus.src.rpm source RPM. I believe<br />
its available via this URL:<br />
<br />
<a href="https://koji.mbox.centos.org/pkgs/packages/kernel-plus/4.18.0/193.19.1.el8_2.centos.plus/src/kernel-plus-4.18.0-193.19.1.el8_2.centos.plus.src.rpm">https://koji.mbox.centos.org/pkgs/packages/kernel-plus/4.18.0/193.19.1.el8_2.centos.plus/src/kernel-plus-4.18.0-193.19.1.el8_2.centos.plus.src.rpm</a><br />
<br />
But as an outsider, I get a 403 error. I can access this page<br />
:<br />
<a href="https://koji.mbox.centos.org/koji/buildinfo?buildID=13615">https://koji.mbox.centos.org/koji/buildinfo?buildID=13615</a><br />
<br />
Which is a bit of a tease... Moving on. Myrebuild script...<br />
<br />
### Kernel patch and rebuild script using the CentOS plus kernel.<br />
sudo dnf --assumeyes --enablerepo=PowerTools install asciidoc audit-libs-devel autoconf automake bash bc binutils binutils-devel bison byacc cmake ctags diffstat diffutils dwarves elfutils elfutils-devel elfutils-libelf-devel expect findutils flex gawk gcc gcc-c++ gdb gettext git glibc-devel gzip hmaccalc hostname intltool java-devel jna kabi-dw libbpf-devel libcap-devel libcap-ng-devel libtool llvm-toolset ltrace m4 make module-init-tools ncurses-devel net-tools newt-devel numactl-devel openssl openssl-devel patch patchutils pciutils-devel perl perl-devel perl-ExtUtils-Embed perl-Fedora-VSP perl-generators pesign pkgconf pkgconf-m4 pkgconf-pkg-config python3-devel python3-docutils redhat-rpm-config rpm-build rpmdevtools rpmlint rpm-sign rsync sh-utils source-highlight strace systemtap tar valgrind valgrind-devel xmlto xz zlib-devel<br />
<br />
## Start the build phase.<br />
mkdir -p ~/rpmbuild/{BUILD,BUILDROOT,RPMS,SOURCES,SPECS,SRPMS}<br />
<br />
# Download the kernel source.<br />
cd ~/rpmbuild/SRPMS/<br />
<br />
# In this case using CentOS kernel-plus v4.18.0-193.14.2.<br />
curl -O <a href="https://mirror.ams1.nl.leaseweb.net/centos-vault/8.2.2004/centosplus/Source/SPackages/kernel-plus-4.18.0-193.14.2.el8_2.centos.plus.src.rpm">https://mirror.ams1.nl.leaseweb.net/centos-vault/8.2.2004/centosplus/Source/SPackages/kernel-plus-4.18.0-193.14.2.el8_2.centos.plus.src.rpm</a><br />
(printf "6693d033cf0c09910505acca2b41bb71307731d539466fb93fb798312e660de5 kernel-plus-4.18.0-193.14.2.el8_2.centos.plus.src.rpm" | sha256sum -c) || (printf "\\n\\nSource RPM did not match the expected hash value.\\n\\n" ; exit 1)<br />
chown $USER:$USER kernel-plus-4.18.0-193.14.2.el8_2.centos.plus.src.rpm<br />
rpm -i kernel-plus-4.18.0-193.14.2.el8_2.centos.plus.src.rpm<br />
<br />
# We need to prep the spec file.<br />
cd ~/rpmbuild/SPECS/<br />
<br />
# This is where the magic happens. We enable SED OPAL block devices.<br />
patch -p1 <<-EOF<br />
diff --git a/kernel-plus.spec b/kernel-plus.spec<br />
index 770a5c8..12331be 100644<br />
--- a/kernel-plus.spec<br />
+++ b/kernel-plus.spec<br />
@@ -1113,6 +1113,9 @@ do<br />
done<br />
%endif<br />
<br />
+# Enable the SED OPAL block device driver.<br />
+sed -i "s/.*CONFIG_BLK_SED_OPAL.*/CONFIG_BLK_SED_OPAL=y/g" *.config<br />
+<br />
cp %{SOURCE42} .<br />
./process_configs.sh -w -c %{name} %{rpmversion}<br />
EOF<br />
<br />
# I decided to give the custom kernel this build ID, so it would be easy<br />
# to spot on a GRUB boot menu.<br />
export BUILDID="_oo00800oo_"<br />
<br />
# And finally to kickoff the build and produce freshly baked kernel RPMs.<br />
rpmbuild -ba --define="buildid $BUILDID" --target=$(uname -m) kernel-plus.spec<br />
###<br />
<br />
The script above was tested and a working kernel built using the "generic/centos8" vagrant <br />
box. <br />
<br />
Finally some GitHub issues which may help.<br />
<br />
<a href="https://github.com/ladar/sedutil/issues/13">https://github.com/ladar/sedutil/issues/13</a><br />
<a href="https://github.com/Drive-Trust-Alliance/sedutil/issues/90">https://github.com/Drive-Trust-Alliance/sedutil/issues/90</a><br />
<br />
And my report to the upstream authorities.<br />
<br />
<a href="https://bugzilla.redhat.com/show_bug.cgi?id=1897014">https://bugzilla.redhat.com/show_bug.cgi?id=1897014</a>
↧