September 2022 - openSUSE Bugs - openSUSE Mailing Lists

[Bug 1188949] New: The wired connection stops connecting the next day after installation
by bugzilla_noreply＠suse.com 06 Feb '23

06 Feb '23

http://bugzilla.opensuse.org/show_bug.cgi?id=1188949 Bug ID: 1188949 Summary: The wired connection stops connecting the next day after installation Classification: openSUSE Product: openSUSE Tumbleweed Version: Current Hardware: x86-64 OS: openSUSE Tumbleweed Status: NEW Severity: Critical Priority: P5 - None Component: Network Assignee: screening-team-bugs(a)suse.de Reporter: vladislavmil(a)mail.ru QA Contact: qa-bugs(a)suse.de Found By: --- Blocker: --- Created attachment 851456 --> http://bugzilla.opensuse.org/attachment.cgi?id=851456&action=edit NetworkManager logs, inxi-Fxz output, video and photo with error demonstration After installation, the wired connection works correctly. The next day, it stops connecting. There is a network connection that lasts indefinitely. Creating new connections does not lead to any result, as does trying to reconnect to an existing connection. It only helps to roll back the system to the backup created immediately after installation. The bug manifests itself on all DE and the next day after installation. During installation, the system configures the wired connection itself and the installation takes place without any errors. The problem manifests itself in openSUSE Leap and Tumbleweed, the wired connection works correctly on other distributions. I tried to install the r8168 driver, and add r8169 to the blacklist, but this does not help. Installing a third-party custom kernel also does not fix the problem. IPv4, IPv6, DNS and the default route disappear from the connection settings. I looked at the Network Manager configs, but there is nothing suspicious there, because of which the problem could manifest itself. -- You are receiving this mail because: You are on the CC list for the bug.

1 46

[Bug 1186027] New: VUL-1: CVE-2021-32917,CVE-2021-32918,CVE-2021-32919,CVE-2021-32920,CVE-2021-32921: Prosody XMPP server advisory 2021-05-12 (multiple vulnerabilities)
by bugzilla_noreply＠suse.com 05 Feb '23

05 Feb '23

http://bugzilla.opensuse.org/show_bug.cgi?id=1186027 Bug ID: 1186027 Summary: VUL-1: CVE-2021-32917,CVE-2021-32918,CVE-2021-32919,CVE-2021- 32920,CVE-2021-32921: Prosody XMPP server advisory 2021-05-12 (multiple vulnerabilities) Classification: openSUSE Product: openSUSE Distribution Version: Leap 15.2 Hardware: Other URL: https://smash.suse.de/issue/284222/ OS: Other Status: NEW Severity: Minor Priority: P5 - None Component: Other Assignee: mvetter(a)suse.com Reporter: gianluca.gabrielli(a)suse.com QA Contact: security-team(a)suse.de Found By: Security Response Team Blocker: --- Prosody security advisory 2021-05-12 ==================================== Project : Prosody XMPP server URL : https://prosody.im/ Date : 2021-05-12 This advisory details 5 new security vulnerabilities discovered in the Prosody.im XMPP server software. All issues are fixed in the 0.11.9 release default configuration. **References** - Release announcement: https://blog.prosody.im/prosody-0.11.9-released/ - Advisory (HTML): https://prosody.im/security/advisory_20210512/ - Advisory (text): https://prosody.im/security/advisory_20210512.txt 1/5: DoS via insufficient memory consumption controls ----------------------------------------------------- CVE : CVE-2021-32918 CVSS : 7.0 (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H/E:F/RL:O/RC:C) CWEs : CWE-400 Affected versions : All versions prior to 0.11.9 Fixed versions : 0.11.9, 0.11 nightly build 130, trunk nightly build 1434 **Description** It was discovered that default settings leave Prosody susceptible to remote unauthenticated denial-of-service (DoS) attacks via memory exhaustion when running under Lua 5.2 or Lua 5.3. Lua 5.2 is the default and recommended Lua version for Prosody 0.11.x series. **Affected configurations** The default configuration is susceptible to this issue. Configurations with stricter settings for stanza size limits, rate limits and garbage collection parameters are at decreased risk from this attack. For more details please review the 'Mitigation' section for recommended values. **Mitigation** Mitigation is possible through configuration changes (on 0.11.7+). All the configuration changes described in this section are applied by default in Prosody 0.11.9. 1) Enable more aggressive garbage collection On Lua 5.2 and 5.3, the garbage collector does not free unused memory fast enough by default. This allowed Prosody's memory usage to grow excessively during certain traffic patterns. It is recommended to set a garbage collection speed of at least 500 in the global section of your configuration file: ``` gc = { speed = 500; } ``` Be aware that this setting may increase CPU usage if the other mitigations in this section are not applied. 2) Enable stricter stanza size limits By default Prosody ships with extremely permissive stanza size limits (up to 10MB). This value was introduced as a way to place a limit on memory usage without affecting legitimate use of the server. However testing demonstrates that the default limit is too high for most deployments. Our recommendation (and the default in 0.11.9) is to adopt the same default size limits that are already enforced by ejabberd, one of the other major XMPP servers on the network. To enable the new limits explicitly, add to the global section of your configuration file the following options: c2s_stanza_size_limit = 256 * 1024 s2s_stanza_size_limit = 512 * 1024 Be aware that reducing limits has the potential to introduce interoperability issues with deployments that do not enforce the same size limits. For example, remote contacts with large avatars. 3) Enable rate limits By default Prosody does not enable any rate limits. However we recommend enabling them for all production and public deployments to ensure fair consumption of resources across all connections. First, ensure that mod_limits is enabled by adding "limits" to your global modules_enabled configuration option: ``` modules_enabled = { ... "limits"; ... } ``` Next, configure the limits: ``` limits = { c2s = { rate = "10kb/s"; }; s2sin = { rate = "30kb/s"; } } ``` **Advice** All public deployments should upgrade to 0.11.9 or apply the above configuration changes. Deployments using nightly builds should upgrade to the latest available builds. **Credits** Many thanks to Travis Burtrum (moparisthebest) for discovering and reporting this issue, and providing a test case. **Commits** - https://hg.prosody.im/trunk/rev/db8e41eb6eff - https://hg.prosody.im/trunk/rev/b0d8920ed5e5 - https://hg.prosody.im/trunk/rev/929de6ade6b6 - https://hg.prosody.im/trunk/rev/63fd4c8465fb - https://hg.prosody.im/trunk/rev/1937b3c3efb5 - https://hg.prosody.im/trunk/rev/3413fea9e6db ---------------------------------------------------------------------------- 2/5: DoS via repeated TLS renegotiation causing excessive CPU consumption ---------------------------------------------------------------------------- CVE : CVE-2021-32920 CVSS : 5.1 (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L/E:H/RL:O/RC:C) CWEs : CWE-400 Affected versions : All versions prior to 0.11.9 Fixed versions : 0.11.9, 0.11 nightly build 130, trunk nightly build 1434 **Description** It was discovered that Prosody does not disable SSL/TLS renegotiation, even though this is not used in XMPP. A malicious client may flood a connection with renegotiation requests to consume excessive CPU resources on the server. Support for disabling renegotiation depends on OpenSSL 1.1.1+ and LuaSec 0.7+. **Affected configurations** The default configuration is susceptible to this issue. **Temporary mitigation** Ensure you have OpenSSL 1.1.1 or higher and LuaSec 0.7 or higher, and set the following ssl option (or add to your existing one if you have one): ``` ssl = { options = { no_renegotiation = true; } } ``` This configuration is applied by default in 0.11.9. **Advice** All public deployments should upgrade to 0.11.9 or apply the above configuration changes. Deployments using nightly builds should upgrade to the latest available builds. **Credits** This flaw was discovered by Kim Alvefur, a member of the Prosody team. **Commits** - https://hg.prosody.im/trunk/rev/55ef50d6cf65 - https://hg.prosody.im/trunk/rev/5a484bd050a7 - https://hg.prosody.im/trunk/rev/aaf9c6b6d18d ----------------------------------------------------------------------- 3/5: Use of timing-dependent string comparison with sensitive values ----------------------------------------------------------------------- CVE : CVE-2021-32921 CVSS : 4.7 (CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:N/E:U/RL:O/RC:U) CWEs : CWE-1254 Affected versions : All versions prior to 0.11.9 Fixed versions : 0.11.9, 0.11 nightly build 130, trunk nightly build 1434 **Description** It was discovered that Prosody does not use a constant-time algorithm for comparing certain secret strings when running under Lua 5.2 or later. This can potentially be used in a timing attack to reveal the contents of secret strings to an attacker. Lua 5.1 utilizes a technique called "string interning", which protected string comparisons from timing attacks. In Lua 5.2 and later versions, strings over 40 bytes in length are excluded from interning. With Prosody running under Lua 5.2, this makes any secret string over 40 bytes in length vulnerable to potential discovery via timing attacks. Note that if a secret string contains non-ASCII (unicode) characters, it may be longer than 40 bytes when encoded as UTF-8 (Prosody's internal encoding) even if it is fewer than 40 characters long. It should be noted that to successfully perform a timing attack, a significant number of failed attempts must typically be made to "guess" at the contents of the secret string. We are not aware of any attempts to exploit this vulnerability (which would likely be noticeable), and no known proof-of-concept exploit exists. **Affected configurations** This flaw affects the following modules: - mod_auth_internal_plain (disabled by default) mod_auth_internal_plain performs a timing-dependent comparison to the user's password if the user's password is longer than 40 bytes. This may allow an attacker to discover a user's password via a timing attack. We do not generally recommend mod_auth_internal_plain for new deployments, and mod_auth_internal_hashed has been the default for Prosody 0.11.x. - mod_muc (disabled by default) mod_muc supports password-protection of MUCs. The password validity check is performed using a timing-dependent comparison, which may allow an attacker to discover the MUC password via a timing attack if the password is longer than 40 bytes. We do not generally recommend using password-protected MUCs. Instead use affiliations to directly grant access to specific JIDs whenever possible. - mod_auth_internal_hashed (enabled by default but not typically vulnerable) mod_auth_internal_hashed has been updated for safety, but it is not vulnerable in the default configuration of Lua 5.2 as the password hashes it compares do not exceed 40 bytes. - mod_dialback (enabled by default but not typically vulnerable) mod_dialback has been updated for safety, but due to the single-use nature of s2s dialback verification strings a timing attack on this module is not believed to be possible, or to grant an attacker any advantage if it were. **Temporary mitigation** mod_auth_internal_plain: we recommend that people upgrade to mod_auth_internal_hashed due to this and also to benefit from its other security properties. mod_muc: use affiliations to grant access to a MUC instead of passwords. If passwords must be used, ensure they are shorter than 40 bytes. Rate limits can greatly lengthen the amount of time required to successfully complete a timing attack. Enable and configure mod_limits. **Advice** All deployments should upgrade to 0.11.9. Deployments using nightly builds should upgrade to the latest available builds. **Credits** This flaw was discovered by Matthew Wild, a member of the Prosody team. The issue with MUC passwords was also previously identified by Robert Gr��sser. **Commits** - https://hg.prosody.im/trunk/rev/c98aebe601f9 - https://hg.prosody.im/trunk/rev/13b84682518e - https://hg.prosody.im/trunk/rev/6f56170ea986 ------------------------------------------------------------------- 4/5: Use of mod_proxy65 is unrestricted in default configuration ------------------------------------------------------------------- CVE : CVE-2021-32917 CVSS : 5.1 (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:L/E:H/RL:O/RC:C) CWEs : CWE-862, CWE-400 Affected versions : All versions prior to 0.11.9 Fixed versions : 0.11.9, 0.11 nightly build 130, trunk nightly build 1434 **Description** mod_proxy65 is a file transfer proxy provided with Prosody to facilitate the transfer of files and other data between XMPP clients. It was discovered that the proxy65 component of Prosody allows open access by default, even if neither of the users have an XMPP account on the local server, allowing unrestricted use of the server's bandwidth. **Affected configurations** The default configuration does not enable mod_proxy65 and is not affected. With mod_proxy65 enabled, all configurations without a 'proxy65_acl' setting configured are affected. **Temporary mitigation** Configure 'proxy65_acl' to a list of XMPP domains that should be allowed to use the file transfer proxy. **Advice** All deployments should upgrade to 0.11.9 and/or configure a 'proxy65_acl' as desired. Deployments using nightly builds should upgrade to the latest available builds. The default behaviour in 0.11.9 allows all local clients to initiate a data stream through the proxy if proxy65_acl is unconfigured. **Credits** This flaw was discovered by the Prosody team. **Commits** - https://hg.prosody.im/trunk/rev/65dcc175ef5b -------------------------------------------------------------- 5/5: Undocumented dialback-without-dialback option insecure -------------------------------------------------------------- CVE : CVE-2021-32919 Affected versions : Prosody 0.10.x, Prosody 0.11.x prior to 0.11.9 Fixed versions : 0.11.9, 0.11 nightly build 130, trunk nightly build 1434 **Description** The undocumented option 'dialback_without_dialback' enabled an experimental feature for server-to-server authentication. A flaw in this feature meant it did not correctly authenticate remote servers, allowing a remote server to impersonate another server when this option is enabled. **Affected configurations** The default configuration is not affected. Configurations with the setting 'dialback_without_dialback' set to true are affected. **Temporary mitigation** Remove or disable the 'dialback_without_dialback' option. **Advice** All deployments should upgrade to 0.11.9 or disable this feature. Deployments using nightly builds should upgrade to the latest available builds. The affected feature has been removed in 0.11.9. **Credits** This flaw was discovered by the Prosody team. **Commits** - https://hg.prosody.im/trunk/rev/6be890ca492e - https://hg.prosody.im/trunk/rev/d0e9ffccdef9 -- You are receiving this mail because: You are on the CC list for the bug.

1 3

[Bug 1203748] New: Laptop very slow in battery mode (Lenovo Thinkpad T470)
by bugzilla_noreply＠suse.com 03 Feb '23

03 Feb '23

https://bugzilla.suse.com/show_bug.cgi?id=1203748 Bug ID: 1203748 Summary: Laptop very slow in battery mode (Lenovo Thinkpad T470) Classification: openSUSE Product: openSUSE Distribution Version: Leap 15.4 Hardware: Other OS: Other Status: NEW Severity: Normal Priority: P5 - None Component: Kernel Assignee: kernel-bugs(a)opensuse.org Reporter: sndirsch(a)suse.com QA Contact: qa-bugs(a)suse.de Found By: --- Blocker: --- For some reason my laptop gets very slow in battery only mode. I notice this when watching videos. It makes it rather unusable when not connecting it to power. It looks like it goes into some deep powersave mode. I would like to avoid this even if this would half the running time in battery mode. Which information do you need? -- You are receiving this mail because: You are on the CC list for the bug.

1 34

[Bug 1202570] New: steam deck - echo freeze | tee -a /sys/powerstate glitches the screen
by bugzilla_noreply＠suse.com 02 Feb '23

02 Feb '23

http://bugzilla.opensuse.org/show_bug.cgi?id=1202570 Bug ID: 1202570 Summary: steam deck - echo freeze | tee -a /sys/powerstate glitches the screen Classification: openSUSE Product: openSUSE Tumbleweed Version: Current Hardware: x86-64 OS: openSUSE Tumbleweed Status: NEW Severity: Normal Priority: P5 - None Component: Kernel Assignee: kernel-bugs(a)opensuse.org Reporter: monkeyboyted(a)yahoo.com QA Contact: qa-bugs(a)suse.de Found By: --- Blocker: --- Suspend seems to work in the latest kernel in kernel:HEAD repo. Software suspends seems to glitch the screen upon re-initialization. Wifi and changing to tty works. lsb_release -a LSB Version: n/a Distributor ID: openSUSE Description: openSUSE Tumbleweed Release: 20220817 Codename: n/a zypper info kernel-default Loading repository data... Reading installed packages... Information for package kernel-default: --------------------------------------- Repository : kernel-repo Name : kernel-default Version : 6.0~rc1-3.1.g7bd57d5 Arch : x86_64 Vendor : obs://build.opensuse.org/Kernel Installed Size : 273.3 MiB Installed : Yes Status : up-to-date Source package : kernel-default-6.0~rc1-3.1.g7bd57d5.nosrc Upstream URL : https://www.kernel.org/ Summary : The Standard Kernel Description : The standard kernel for both uniprocessor and multiprocessor systems. Source Timestamp: 2022-08-16 07:58:42 +0000 GIT Revision: 7bd57d59ba8c141be35bb5189487a899d04e76ec GIT Branch: master -- You are receiving this mail because: You are on the CC list for the bug.

1 8

[Bug 1202566] New: Steam deck wakes up from hibernate with max brightness
by bugzilla_noreply＠suse.com 02 Feb '23

02 Feb '23

http://bugzilla.opensuse.org/show_bug.cgi?id=1202566 Bug ID: 1202566 Summary: Steam deck wakes up from hibernate with max brightness Classification: openSUSE Product: openSUSE Tumbleweed Version: Current Hardware: x86-64 OS: openSUSE Tumbleweed Status: NEW Severity: Normal Priority: P5 - None Component: Kernel Assignee: kernel-bugs(a)opensuse.org Reporter: monkeyboyted(a)yahoo.com QA Contact: qa-bugs(a)suse.de Found By: --- Blocker: --- echo 0x1e | sudo tee -a /sys/module/drm/parameters/debug [sudo] password for root: 0x1e -> echo disk | sudo tee -a /sys/power/state disk ~> dmesg > dmesg-suspend-from-disk-with-0x1e-enabled lsb_release -a LSB Version: n/a Distributor ID: openSUSE Description: openSUSE Tumbleweed Release: 20220817 Codename: n/a Linux steamdeck 6.0.0-rc1-1.gd997b88-default #1 SMP PREEMPT_DYNAMIC Thu Aug 18 14:44:01 UTC 2022 (d997b88) x86_64 x86_64 x86_64 GNU/Linux https://build.opensuse.org/project/show/home:tiwai:bsc1201418 -- You are receiving this mail because: You are on the CC list for the bug.

1 10

[Bug 1203946] New: btrfs scrub maintenance automatically starts a balance before starting scrub
by bugzilla_noreply＠suse.com 02 Feb '23

02 Feb '23

http://bugzilla.opensuse.org/show_bug.cgi?id=1203946 Bug ID: 1203946 Summary: btrfs scrub maintenance automatically starts a balance before starting scrub Classification: openSUSE Product: openSUSE Tumbleweed Version: Current Hardware: Other OS: Other Status: NEW Severity: Normal Priority: P5 - None Component: Other Assignee: screening-team-bugs(a)suse.de Reporter: jshand2013(a)gmail.com QA Contact: qa-bugs(a)suse.de Found By: --- Blocker: --- i have stumbled across a small issue where when btrfsmaintenance stakes place with scrub on monthly basis, scrub maintenance seems to start a full system balance before doing the scrub itself... is this by design? before i changed doing a balance once a week to monthly, this issue happens, usually on the 1st of the month. -- You are receiving this mail because: You are on the CC list for the bug.

1 4

[Bug 1182554] New: cannot start greetd.service .... missing PAM module
by bugzilla_noreply＠suse.com 02 Feb '23

02 Feb '23

https://bugzilla.suse.com/show_bug.cgi?id=1182554 Bug ID: 1182554 Summary: cannot start greetd.service .... missing PAM module Classification: openSUSE Product: openSUSE Tumbleweed Version: Current Hardware: x86-64 OS: openSUSE Tumbleweed Status: NEW Severity: Enhancement Priority: P5 - None Component: Other Assignee: screening-team-bugs(a)suse.de Reporter: mcepl(a)suse.com QA Contact: qa-bugs(a)suse.de CC: denis.kondratenko(a)suse.com, mcepl(a)suse.com, opensuse(a)trummer.xyz, stdden(a)opensuse.org Found By: --- Blocker: --- Created attachment 846366 --> https://bugzilla.suse.com/attachment.cgi?id=846366&action=edit journalctl -xb output When trying to replace gdm with greetd following the steps on https://en.opensuse.org/Portal:OpenSUSEway, greetd fails to start and journal shows: �no 22 09:38:10 stitny greetd[2347]: error: PAM 'greetd' service missing When looking around, there is the file /usr/etc/pam.d/greetd with this content: ~@stitny$ cat /usr/etc/pam.d/greetd #%PAM-1.0 auth requisite pam_nologin.so auth include common-auth account include common-account password include common-password session required pam_loginuid.so session optional pam_keyinit.so force revoke session include common-session ~@stitny$ but I have still /etc/pam.d/ as well and there is no greetd file. Which one is used? -- You are receiving this mail because: You are on the CC list for the bug.

1 14

[Bug 1165273] New: zathura-plugin-pdf-mupdf not loaded due to error.
by bugzilla_noreply＠novell.com 31 Jan '23

31 Jan '23

http://bugzilla.opensuse.org/show_bug.cgi?id=1165273 Bug ID: 1165273 Summary: zathura-plugin-pdf-mupdf not loaded due to error. Classification: openSUSE Product: openSUSE Tumbleweed Version: Current Hardware: x86-64 OS: Other Status: NEW Severity: Normal Priority: P5 - None Component: Other Assignee: screening-team-bugs(a)suse.de Reporter: dracaphallus(a)gmail.com QA Contact: qa-bugs(a)suse.de Found By: --- Blocker: --- User-Agent: Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/80.0.3987.100 Safari/537.36 Build Identifier: Running zathura after the installation of zathura-plugin-pdf-mupdf the following error output is produced: ``` error: Could not load plugin '/usr/lib64/zathura/libpdf-mupdf.so' (/usr/lib64/zathura/libpdf-mupdf.so: undefined symbol: jpeg_resync_to_restart). ``` >From the error output I'd deduced that the package was packaged wrong and therefore produced the error mentioned above. It has the plugin from working as intended by making zathura incapable of opening the supported document formats (PDF, ePub, and OpenXPS). This issue was reported 14 days ago using the comment section of openSUSE:Factory, while it was reported 8 days ago in the comment section of `X11:Utilities` with no response. This has lead me to file a bug report in OpenSUSE Bugzilla with the hope of getting this issue resolved. Reproducible: Always Steps to Reproduce: 1. Installing zathura-plugin-pdf-mupdf using zypper. 2. Running zathura through terminal. Actual Results: The following error was produced: ``` error: Could not load plugin '/usr/lib64/zathura/libpdf-mupdf.so' (/usr/lib64/zathura/libpdf-mupdf.so: undefined symbol: jpeg_resync_to_restart). ``` Expected Results: Zathura being able to open the supported document formats (PDF, ePub, and OpenXPS). -- You are receiving this mail because: You are on the CC list for the bug.

2 2

[Bug 1200564] New: io_uring instability on ppc64
by bugzilla_noreply＠suse.com 30 Jan '23

30 Jan '23

https://bugzilla.suse.com/show_bug.cgi?id=1200564 Bug ID: 1200564 Summary: io_uring instability on ppc64 Classification: openSUSE Product: openSUSE Tumbleweed Version: Current Hardware: PowerPC-64 OS: Other Status: NEW Severity: Major Priority: P5 - None Component: Kernel Assignee: kernel-bugs(a)opensuse.org Reporter: dmueller(a)suse.com QA Contact: qa-bugs(a)suse.de Found By: --- Blocker: --- when using aio=io_uring in qemu for KVM virtual machine guests on a SLE15SP4 or a opensuse tumbleweed host kernel on power8 or power9 machine, we have very rapid I/O corruption (within milliseconds-seconds) in the guest. on all other architectures things work perfectly fine. -- You are receiving this mail because: You are on the CC list for the bug.

1 15

[Bug 1202786] New: RPM scriptlets not compatible with obs_scm
by bugzilla_noreply＠suse.com 29 Jan '23

29 Jan '23

https://bugzilla.suse.com/show_bug.cgi?id=1202786 Bug ID: 1202786 Summary: RPM scriptlets not compatible with obs_scm Classification: openSUSE Product: openSUSE Distribution Version: Leap 15.4 Hardware: x86-64 OS: openSUSE Leap 15.4 Status: NEW Severity: Minor Priority: P5 - None Component: Development Assignee: screening-team-bugs(a)suse.de Reporter: georg.pfuetzenreuter(a)suse.com QA Contact: qa-bugs(a)suse.de Found By: --- Blocker: --- Hello! I am unsure whether upstream RPM would be a better fit for this issue, however since it is partially related to behavior introduced by SUSE's obs_scm, I decided to post this here: Certain RPM macros make use of %pre/%post scriptlets, invoked using `%pre/%post -f <file>` - one good example which I will use here are the `sysuser` macros, which generate a dynamic %pre scriptlet during build. This works fine in traditional spec files, however if obs_scm with is used as the single source, following the instructions in https://openbuildservice.org/help/manuals/obs-user-guide/cha.obs.best-pract…, this fails, as RPM tries to locate the `%pre -f` scriptlet in a bogus directory: ``` Source0: _service %prep %setup -q -n %{_sourcedir}/%{appname}-%{version} -T -D %build %sysusers_generate_pre %{SOURCE1} %{appname} system-user-%{appname}.conf RPM build errors: Could not open Postin file /home/abuild/rpmbuild/BUILD/home/abuild/rpmbuild/SOURCES/Limnoria-20220814/limnoria.pre: No such file or directory ``` Notice how it prepends the SOURCES plus the BUILD directory in front of the .pre file, instead of simply referencing the working directory, which would be the build directory (where the .pre file was correctly placed in by the %sysusers_generate_pre macro, as confirmed by inspecting the broken build environment). A workaround is to use the obs recompress service to get a .tar.gz source, which one then uses a "traditional" `%setup` call for. However, I deem this should not be necessary just to get `%pre/post -f` scriptlets to be referenced from the correct working directory. A second work around is to dump the contents of the generated .pre file and pasting them directly into the section without the `-f <file>` call - but that defeats the point of having a generated/dynamic scriptlet. I attempted to supply the full path to `-f` as well, and confirmed it's location, however it would still prepend the bogus directory. You can find a sample .spec file here, however it may change at will, since I am actively working on it: https://build.opensuse.org/package/view_file/home:crameleon:LibertaCasa/pyt… Best, Georg -- You are receiving this mail because: You are on the CC list for the bug.

1 8