This allows us to protect the host system a little bit more from the
host system by decoupling all namespaces.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Since we now mount everything in a new namespace, there is no need to
clean up ourselves. This will be done when the last process leaves the
namespace.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
- Everytime an update has been done on squid the access.log file has been replaced with an
empty file, losing whatever messages have been in the log.
- This has been the case since squid was implemented in IPFire.
- Update of rootfile to comment out var/log/squid/access.log
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
For details see:
https://dlcdn.apache.org/httpd/CHANGES_2.4.61
"Changes with Apache 2.4.61
*) SECURITY: CVE-2024-39884: Apache HTTP Server: source code
disclosure with handlers configured via AddType (cve.mitre.org)
A regression in the core of Apache HTTP Server 2.4.60 ignores
some use of the legacy content-type based configuration of
handlers. "AddType" and similar configuration, under some
circumstances where files are requested indirectly, result in
source code disclosure of local content. For example, PHP
scripts may be served instead of interpreted.
Users are recommended to upgrade to version 2.4.61, which fixes
this issue."
Signed-off-by: Matthias Fischer <matthias.fischer@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
- Update from version 4.20.1 to 4.20.2
- Update of rootfile for both x86_64 and aarch64
- After doing a grep into the config directories I realised that the xxxMACHINExxx phrase
is only added into rootfiles in the main common or package directories and not in the
x86_64 and aarch64
- In the past I have submitted the samba rootfile with x86_64 replaced by xxxMACHINExxx.
It seems to have worked, so the replacement probably occurs even in the architecture
specific directories but it doesn't need to be used there as the directory is clearly
only for that one architecture.
- Changelog
4.20.2
* BUG 15662: vfs_widelinks with DFS shares breaks case insensitivity.
* BUG 13213: Samba build is not reproducible.
* BUG 15569: ldb qsort might r/w out of bounds with an intransitive compare
function.
* BUG 15625: Many qsort() comparison functions are non-transitive, which can
lead to out-of-bounds access in some circumstances.
* BUG 15638: Need to change gitlab-ci.yml tags in all branches to avoid CI
bill.
* BUG 15654: We have added new options --vendor-name and --vendor-patch-
revision arguments to ./configure to allow distributions and packagers to
put their name in the Samba version string so that when debugging Samba the
source of the binary is obvious.
* BUG 15665: CTDB RADOS mutex helper misses namespace support.
* BUG 13019: Dynamic DNS updates with the internal DNS are not working.
* BUG 14981: netr_LogonSamLogonEx returns NR_STATUS_ACCESS_DENIED with
SysvolReady=0.
* BUG 15412: Anonymous smb3 signing/encryption should be allowed (similar to
Windows Server 2022).
* BUG 15573: Panic in dreplsrv_op_pull_source_apply_changes_trigger.
* BUG 15620: s4:nbt_server: does not provide unexpected handling, so winbindd
can't use nmb requests instead cldap.
* BUG 15642: winbindd, net ads join and other things don't work on an ipv6
only host.
* BUG 15659: Segmentation fault when deleting files in vfs_recycle.
* BUG 15664: Panic in vfs_offload_token_db_fetch_fsp().
* BUG 15666: "client use kerberos" and --use-kerberos is ignored for the
machine account.
* BUG 15435: Regression DFS not working with widelinks = true.
* BUG 15633: samba-gpupdate - Invalid NtVer in netlogon_samlogon_response.
* BUG 15653: idmap_ad creates an incorrect local krb5.conf in case of trusted
domain lookups.
* BUG 15660: The images don't build after the git security release and CentOS
8 Stream is EOL.
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
The build system defaults to building the library for the host system
and therefore uses instructions that might not be available on the
target system.
This patch changes that we will build the library so that it will choose
the most optimised functions at runtime.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
This enables some DoS protection using SYNPROXY which will complete a
SYN handshake with the client before the connection is being forwarded.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
This should never cause any problems, but will cause that certain more
complicated featured like SYNPROXY won't work.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
The SSH daemon has been split into a listener and session daemon to have
a smaller attack vector since the listener does not need to implement
the SSH protocol.
In order to keep SSH working, we need to ship the session daemon, too.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
- Update from version 9.7p1 to 9.8p1
- Update of rootfile
- Changelog
9.8p1
-There is a fix for CVE-2024-6387
-The number of changes is too large to show all here. As well as the CVE fix and
another security related fix there are a log of bug fixes as well. The details can
seen at https://www.openssh.com/txt/release-9.8
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Excerpt from changelog:
"7.0.6 -- 2024-06-26
Security #7042: defrag: id reuse can lead to invalid reassembly (7.0.x backport)(CVE 2024-37151)
Security #7105: http2: oom from duplicate headers (7.0.x backport)
Security #7033: http/range: segv when http.memcap is reached (7.0.x backport)
Security #6988: modbus: txs without responses are never freed (7.0.x backport)
Bug #7107: packet: app-layer-events incorrectly used on recycled packets (7.0.x backport)
Bug #7064: util/radix-tree: Possible dereference of nullptr in case of unsuccess allocation of memory for node (7.0.x backport)
Bug #7063: smtp/mime: data command rejected by pipelining server does not reset data mode (7.0.x backport)
Bug #7060: smtp: split name logged as 2 names (7.0.x backport)
Bug #7050: af-packet: failure to start up on many threads plus high load (7.0.x backport)
Bug #7043: Crasher in HTTP chunked / StreamingBuffer (7.0.x backport)
Bug #7038: pcap/log: MacOS rotates file well before limit is reached (7.0.x backport)
Bug #7035: time: in offline mode, time can stay behind at pcap start (7.0.x backport)
Bug #7023: unix-socket: iface-bypassed-stat crash (7.0.x backport)
Bug #7021: unix-socket: hostbit commands don't properly release host (7.0.x backport)
Bug #7015: rust: build with rust 1.78 with slice::from_raw_parts now requiring the pointer to be non-null (7.0.x backport)
Bug #6990: tls.random buffers don't work as expected (7.0.x backport)
Bug #6986: iprep: rule with '=,0' can't match (7.0.x backport)
Bug #6975: detect: log relevant frames app-layer metdata (7.0.x backport)
Bug #6950: decode/ppp: decoder.event.ppp.wrong_type on valid packet (7.0.x backport)
Bug #6897: detect/port: upper boundary ports are not correctly handled (7.0.x backport)
Bug #6895: detect/port: port grouping does not happen correctly if gap between a single and range port (7.0.x backport)
Bug #6862: Lightweight rules profiling: crash when profiling ends (7.0.x backport)
Bug #6848: alerts: wrongly using tx id 0 when there is no tx (7.0.x backport)
Bug #6845: coverity: warning in port grouping code (7.0.x backport)
Bug #6844: detect/port: port ranges are incorrect when a port is single as well as a part of range (7.0.x backport)
Bug #6690: Ethernet src should match src ip (7.0.x backport)
Bug #6520: detect-engine/port: recursive DetectPortInsert calls are expensive (7.0.x backport)
Optimization #6830: detect/port: port grouping is quite slow in worst cases (7.0.x backport)
Optimization #6829: detect/port: PortGroupWhitelist fn takes a lot of processing time (7.0.x backport)
Feature #7010: JA4 support for TLS and QUIC (7.0.x backport)
Feature #6557: Capability to have rules profiling on pcap run (7.0.x backport)
Documentation #6910: userguide: document how to verify tar.gz signature (7.0.x backport)
Documentation #6687: docs: port userguide build instruction changes from master-6.0.x (7.0.x backport)
Documentation #6601: docs: update eBPF installation instructions (7.0.x backport)"
Signed-off-by: Matthias Fischer <matthias.fischer@ipfire.org>
Reviewed-by: Michael Tremer <michael.tremer@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
For details see:
https://www.nano-editor.org/news.php
"2024 May 1 - GNU nano 8.0 "Grus grus"
By default ^F is bound to starting a forward search, and ^B to
starting a backward search, while M-F and M-B repeat the search
in the corresponding direction. (See the documentation if you
want the old bindings back.)
Command-line option --modernbindings (-/) makes ^Q quit, ^X cut,
^C copy, ^V paste, ^Z undo, ^Y redo, ^O open a file, ^W write a file,
^R replace, ^G find again, ^D find again backwards, ^A set the mark,
^T jump to a line, ^P show the position, and ^E execute.
Above modern bindings are activated also when the name of
nano's executable (or a symlink to it) starts with the letter "e".
To open a file at a certain line number, one can now use also
nano filename:number, besides nano +number filename.
<Alt+Home> and <Alt+End> put the cursor on the first and last
row in the viewport, while retaining the horizontal position.
When the three digits in an #RGB color code are all the same,
the code is mapped to the xterm grey scale, giving access to
fourteen levels of grey instead of just four.
For easier access, M-" is bound to placing/removing an anchor,
and M-' to jumping to the next anchor.
Whenever an error occurs, the keystroke buffer is cleared, thus
stopping the execution of a macro or a string bind.
The mousewheel scrolls the viewport instead of moving the cursor."
Signed-off-by: Matthias Fischer <matthias.fischer@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
