Development April 2024

development@lists.ipfire.org

10 participants
33 discussions

[PATCH] Stop unbound-dhcp-leases-bridge from continually restarting unbound
by Nick Howitt 30 Apr '24

30 Apr '24

Hi all, **** This is a repost of an email incorrectly submitted. **** Please bear with me as I am new to IPFire and not at all used to this style of development, so any guidance would be appreciated. There is a bug, https://bugzilla.ipfire.org/show_bug.cgi?id=13254, where unbound-dhcp-leases-bridge restarts unbound a lot when it is totally unnecessary. It appears to be because when a lease is renewed, the script gets triggered, created a new /etc/unbound/dhcp-leases.conf then restarts Unbound to read the file. Generally this seems to be unnecessary as, with a renewed lease, the old and new /etc/unbound/dhcp-leases.conf files are the same. With 5 leases in that file (one for a machine not active) I am getting 3-4 restarts an hour of Unbound when I have a min/max lease time of 60/120min. Looking at the code, it is fairly easy to fix. The current code creates a temp file then copies it into place then restarts unbound. All I am doing is doing a file comparison before the copy and skipping the restart if the files are the same. There were a couple of gotchas because setting the file attributes and copying the file were done inside the "with" block for generating the temporary file. This meant a file comparison always returned False as the temp file was still open and so never the same is the old file. I moved those two statements outside the "with". This forced me to change the fchmod to chmod. It could be argued that the file copy should not be done if the files are not different, but it would take an extra "if" and you'd have to remember to delete the temp file. If required, I can make that change. Also, one small thing I noticed once is that the old and new dhcp-leases.conf files could occasionally contain the same leases but in a different order. I have been unable to reproduce, but to sidestep it, instead of stepping through the leases variable directly, I sort it and step through that. It should make the resulting file completely deterministic and make the file comparison more effective. My patch is: From 73873d4944944a2f02317a73074a6894726f36f7 Mon Sep 17 00:00:00 2001 From: Nick Howitt <nick(a)howitts.co.uk> Date: Sun, 24 Mar 2024 15:17:19 +0000 Subject: [PATCH] Stop unbound-dhcp-leases-bridge from restarting unbound every time it write dhcp-leases.conf as it is very often unchanged and does not require a restart. --- config/unbound/unbound-dhcp-leases-bridge | 28 +++++++++++++++-------- 1 file changed, 19 insertions(+), 9 deletions(-) diff --git a/config/unbound/unbound-dhcp-leases-bridge b/config/unbound/unbound-dhcp-leases-bridge index e9f022aff..d22772066 100644 --- a/config/unbound/unbound-dhcp-leases-bridge +++ b/config/unbound/unbound-dhcp-leases-bridge @@ -22,6 +22,7 @@ import argparse import datetime import daemon +import filecmp import functools import ipaddress import logging @@ -516,26 +517,35 @@ class UnboundConfigWriter(object): def update_dhcp_leases(self, leases): # Write out all leases - self.write_dhcp_leases(leases) + if self.write_dhcp_leases(leases): - log.debug("Reloading Unbound...") + log.debug("Reloading Unbound...") - # Reload the configuration without dropping the cache - self._control("reload_keep_cache") + # Reload the configuration without dropping the cache + self._control("reload_keep_cache") + + else: + + log.debug("Not reloading Unbound. Leases not changed.") def write_dhcp_leases(self, leases): log.debug("Writing DHCP leases...") with tempfile.NamedTemporaryFile(mode="w", delete=False) as f: - for l in leases: + for l in sorted(leases): for rr in l.rrset: f.write("local-data: \"%s\"\n" % " ".join(rr)) - # Make file readable for everyone - os.fchmod(f.fileno(), stat.S_IRUSR|stat.S_IWUSR|stat.S_IRGRP|stat.S_IROTH) + filecmp.clear_cache() + RequireUnboundReload = not filecmp.cmp(f.name, self.path, shallow=False) + + # Make file readable for everyone + os.chmod(f.name, stat.S_IRUSR|stat.S_IWUSR|stat.S_IRGRP|stat.S_IROTH) + + # Move the file to its destination + os.rename(f.name, self.path) - # Move the file to its destination - os.rename(f.name, self.path) + return RequireUnboundReload def _control(self, *args): command = ["unbound-control"] -- 2.43.0 I have this locally on a branch on my firewall. If it is OK, do I need some sort of rights to push my branch and then how I get it merged? Regards, Nick

2 15

[PATCH 1/3] unbound-dhcp-leases-bridge: Implement atomic file replacement
by Michael Tremer 30 Apr '24

30 Apr '24

This change no longer renames the file, but removes the old link and creates a new link for the temporary file. That helps us to jump out of the code at any point without worrying about cleaning up the temporary file. Signed-off-by: Michael Tremer <michael.tremer(a)ipfire.org> --- config/unbound/unbound-dhcp-leases-bridge | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/config/unbound/unbound-dhcp-leases-bridge b/config/unbound/unbound-dhcp-leases-bridge index e9f022aff..5d5696af0 100644 --- a/config/unbound/unbound-dhcp-leases-bridge +++ b/config/unbound/unbound-dhcp-leases-bridge @@ -526,16 +526,22 @@ class UnboundConfigWriter(object): def write_dhcp_leases(self, leases): log.debug("Writing DHCP leases...") - with tempfile.NamedTemporaryFile(mode="w", delete=False) as f: + with tempfile.NamedTemporaryFile(mode="w") as f: for l in leases: for rr in l.rrset: f.write("local-data: \"%s\"\n" % " ".join(rr)) + # Flush the file + f.flush() + # Make file readable for everyone os.fchmod(f.fileno(), stat.S_IRUSR|stat.S_IWUSR|stat.S_IRGRP|stat.S_IROTH) + # Remove the old file + os.unlink(self.path) + # Move the file to its destination - os.rename(f.name, self.path) + os.link(f.name, self.path) def _control(self, *args): command = ["unbound-control"] -- 2.39.2

2 7

[PATCH] apache: Fixes bug13656 - Add delay between stop & start of restart command.
by Adolf Belka 29 Apr '24

29 Apr '24

- The change of the apache initscript in CU181 made the restart comand a stop followed by a start. From bug13656 it looks like when the start is run the pid file may not yet be fully removed and so apache is not started, although the pid is being removed. - This patch checks if the pid file is still present every second up to 10 seconds. Once the pid file is gone then the stop command is completed and the initscript moves to the start command. - Rather than apply a fixed delay of 2 or 3 or 4 seconds I used a while loop to check every second if the file is still present. If at the end of 10 seconds it is still present then something went wrong with the pid removal. - I have tested this patch on my vm system and it worked but it would be good to be reviewed to make sure that it is a reasonable approach that has been used and if required changed in whatever way makes the best sense. Fixes: Bug13656 Tested-by: Adolf Belka <adolf.belka(a)ipfire.org> Signed-off-by: Adolf Belka <adolf.belka(a)ipfire.org> --- src/initscripts/system/apache | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-) diff --git a/src/initscripts/system/apache b/src/initscripts/system/apache index 18eb86e2f..087e4084e 100644 --- a/src/initscripts/system/apache +++ b/src/initscripts/system/apache @@ -2,7 +2,7 @@ ############################################################################### # # # IPFire.org - A linux based firewall # -# Copyright (C) 2007-2022 IPFire Team <info(a)ipfire.org> # +# Copyright (C) 2007-2024 IPFire Team <info(a)ipfire.org> # # # # This program is free software: you can redistribute it and/or modify # # it under the terms of the GNU General Public License as published by # @@ -87,6 +87,16 @@ case "$1" in stop) boot_mesg "Stopping Apache daemon..." /usr/sbin/apachectl -k stop + COUNTER=0 + while [ -e /var/run/httpd.pid ] + do + sleep 1 + (( COUNTER++ )) + if [ $COUNTER -eq 10 ]; then + boot_mesg "pid not removed after 10 seconds" + break + fi + done evaluate_retval ;; -- 2.44.0

3 6

[PATCH] mympd: update to 14.1.2
by Arne Fitzenreiter 29 Apr '24

29 Apr '24

This is a small bugfix release. Changelog: Fix: Output enabled state is bool type Fix: Add missing sort parameters to home icon for search Fix: Set default stream port if stream uri is defined Signed-off-by: Arne Fitzenreiter <arne_f(a)ipfire.org> --- lfs/mympd | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/lfs/mympd b/lfs/mympd index bb7215ba1..012f867da 100644 --- a/lfs/mympd +++ b/lfs/mympd @@ -26,7 +26,7 @@ include Config SUMMARY = Webfrontend for Music Player Daemon -VER = 14.1.0 +VER = 14.1.2 THISAPP = myMPD-$(VER) DL_FILE = $(THISAPP).tar.gz @@ -34,7 +34,7 @@ DL_FROM = $(URL_IPFIRE) DIR_APP = $(DIR_SRC)/$(THISAPP) TARGET = $(DIR_INFO)/$(THISAPP) PROG = mympd -PAK_VER = 4 +PAK_VER = 5 DEPS = mpd libmpdclient @@ -48,7 +48,7 @@ objects = $(DL_FILE) $(DL_FILE) = $(DL_FROM)/$(DL_FILE) -$(DL_FILE)_BLAKE2 = 621ecc2b9ecf78c606005120837480b1bccc92fd3b9c201781e719ea524fc689b58f587f59fd388b73a0a1e63befce951e9182ee865e625f83a175a61572759b +$(DL_FILE)_BLAKE2 = 59a5d62ed1e91d422f091f331bd935b0988715fd641b81f48e19a37694cc14da2c78d4563bb149767afd873bcebdfaccfd984b63f45a6bb321214eebb3807ddf install : $(TARGET) -- 2.39.2

1 0

[PATCH] initscripts: Correctly wait for Apache2 to terminate
by Michael Tremer 26 Apr '24

26 Apr '24

This is achieved by telling killproc which PIDs to wait for. Signed-off-by: Michael Tremer <michael.tremer(a)ipfire.org> --- src/initscripts/system/apache | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/src/initscripts/system/apache b/src/initscripts/system/apache index 18eb86e2f..e7a62097e 100644 --- a/src/initscripts/system/apache +++ b/src/initscripts/system/apache @@ -22,6 +22,8 @@ . /etc/sysconfig/rc . $rc_functions +PIDFILE="/var/run/httpd.pid" + generate_certificates() { if [ ! -f "/etc/httpd/server.key" ]; then boot_mesg "Generating HTTPS RSA server key (this will take a moment)..." @@ -86,8 +88,7 @@ case "$1" in stop) boot_mesg "Stopping Apache daemon..." - /usr/sbin/apachectl -k stop - evaluate_retval + killproc /usr/sbin/httpd ;; restart) -- 2.39.2

1 0

[PATCH] sqlite: Update to version 3450300
by Adolf Belka 26 Apr '24

26 Apr '24

- Update from version 3450200 to 3450300 - Update of rootfile not required - Changelog 3450300 Fix a long-standing bug (going back to version 3.24.0) that might (rarely) cause the "old.*" values of an UPDATE trigger to be incorrect if that trigger fires in response to an UPSERT. Forum post 284955a3cd454a15. Fix a bug in sum() that could cause it to return NULL when it should return Infinity. Forum post 23b8688ef4. Other trifling corrections and compiler warning fixes that have come up since the previous patch release. See the timeline for details. Signed-off-by: Adolf Belka <adolf.belka(a)ipfire.org> --- lfs/sqlite | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/lfs/sqlite b/lfs/sqlite index 922b303b7..421e32dd2 100644 --- a/lfs/sqlite +++ b/lfs/sqlite @@ -24,7 +24,7 @@ include Config -VER = 3450200 +VER = 3450300 THISAPP = sqlite-autoconf-$(VER) DL_FILE = $(THISAPP).tar.gz @@ -40,7 +40,7 @@ objects = $(DL_FILE) $(DL_FILE) = $(DL_FROM)/$(DL_FILE) -$(DL_FILE)_BLAKE2 = 7bc225167a6792a35c2e7a324fe9bff1a1540a01b0fd04197d023f113a368ba6034235045281c7653abdd2ea67aa10a5ed19b024d1d25bdb7232533e25dfb991 +$(DL_FILE)_BLAKE2 = 24e47dcfb8fb1dbd0055d48cfc540884439e9443b1b79700e867326feb60b340a9b5d5bdb83eeca210b47251e61b706ec67eb1b84e59b11bae6c76271b1030bf install : $(TARGET) -- 2.44.0

2 3

Re: Enable eBPF XDP/TC kernel feature for IPFire
by Michael Tremer 25 Apr '24

25 Apr '24

Hello, Please make sure you keep the list copied. > On 24 Apr 2024, at 20:55, Vincent Li <vincent.mc.li(a)gmail.com> wrote: > > On Wed, Apr 24, 2024 at 8:28 AM Michael Tremer > <michael.tremer(a)ipfire.org> wrote: >> >> Hello, >> >>> On 19 Apr 2024, at 02:17, Vincent Li <vincent.mc.li(a)gmail.com> wrote: >>> >>> On Thu, Apr 18, 2024 at 2:13 PM Michael Tremer >>> <michael.tremer(a)ipfire.org> wrote: >>>> >>>> Hello Vincent, >>>> >>>>> On 18 Apr 2024, at 16:21, Vincent Li <vincent.mc.li(a)gmail.com> wrote: >>>>> >>>>> On Thu, Apr 18, 2024 at 1:57 AM Michael Tremer >>>>> <michael.tremer(a)ipfire.org> wrote: >>>>>> >>>>>> Hello, >>>>>> >>>>>>> On 17 Apr 2024, at 23:36, Vincent Li <vincent.mc.li(a)gmail.com> wrote: >>>>>>> >>>>>>> On Wed, Apr 17, 2024 at 9:07 AM Michael Tremer >>>>>>> <michael.tremer(a)ipfire.org> wrote: >>>>>>>> >>>>>>>> Hello Vincent, >>>>>>>> >>>>>>>>> On 10 Apr 2024, at 19:01, Vincent Li <vincent.mc.li(a)gmail.com> wrote: >>>>>>>>> >>>>>>>>> On Wed, Apr 10, 2024 at 8:17 AM Peter Müller <peter.mueller(a)ipfire.org> wrote: >>>>>>>>>> >>>>>>>>>> Hello Vincent, >>>>>>>>>> >>>>>>>>>> thank you for your e-mail and the proposal. >>>>>>>>>> >>>>>>>>>>> Hi Adolf, >>>>>>>>>>> >>>>>>>>>>> Please see my reply inline >>>>>>>>>>> >>>>>>>>>>> On Wed, Apr 10, 2024 at 2:04 AM Adolf Belka <adolf.belka(a)ipfire.org> wrote: >>>>>>>>>>>> >>>>>>>>>>>> Hi Vincent, >>>>>>>>>>>> >>>>>>>>>>>> I am not very familiar at all with this type of stuff but one thing that I noticed is that in the image you provided a link to, the XDP section has a line labelled XDP_TX which completely bypasses the whole Netfilter section which doesn't seem to be a good idea to me. >>>>>>>>>>>> >>>>>>>>>>> XDP_TX is to redirect the packet out after processing the packet at >>>>>>>>>>> XDP stage, yes, netfilter will not see these packets. >>>>>>>>>>> for example for DDoS SYN flood attack scenario, when the SYN packet >>>>>>>>>>> is received, XDP program can generate SYN+ACK with syncookie and send >>>>>>>>>>> the SYN+ACK out, netfilter/Linux tcp stack knows nothing about it, >>>>>>>>>>> which actually saves host CPU cycles to process the SYN in >>>>>>>>>>> netfilter/TCP stack, which is actually good thing. >>>>>>>>>>> >>>>>>>>>>> Also, XDP_DROP, XDP_PASS, XDP_TX action is depending on the XDP >>>>>>>>>>> program attached to the network interface, so it is the XDP program >>>>>>>>>>> author decide what to do with the packet, if no XDP program attached >>>>>>>>>>> to the network interface, everything works as usual, no interference >>>>>>>>>>> from XDP. >>>>>>>>>> >>>>>>>>>> If my understanding of this is correct, then this would lead to the exact >>>>>>>>>> opposite of what IPFire is designed to do. Rather than having packets >>>>>>>>>> processed below any level of operating system influence, the objective of >>>>>>>>>> IPFire in particular and firewalls in general is to control network traffic, >>>>>>>>>> which inherently requires thorough visibility on it. >>>>>>>>> >>>>>>>>> Kernel still has the traffic statistics processed by XDP program and >>>>>>>>> store in eBPF maps so the user space program can query and view. you >>>>>>>>> can still view XDP as part of the firewall except it processes packets >>>>>>>>> early at the driver layer for efficiency. >>>>>>>> >>>>>>>> I would like to understand what your need is to use XDP. >>>>>>>> >>>>>>>> As Peter has stated, your system will pass packets with this, but 90% of the features that IPFire has won’t work any more: >>>>>>>> >>>>>>>> * Connection Tracking won’t be up to date >>>>>>>> * QoS won’t be able to categorise packets correctly and won’t be able to do its job >>>>>>>> * The IPS won’t be able to inspect any data >>>>>>>> >>>>>>> >>>>>>> I think I did not explain the XDP use case clearly, for now, most XDP >>>>>>> use cases and particularly my use case is to do DDoS protection at >>>>>>> the earliest packet receiving path in high efficiency since XDP works >>>>>>> inside the network driver. the things you mentioned above would work >>>>>>> fine even if the packets go through XDP program, yes XDP program can >>>>>>> drop, modify, reflect the packet, but in non-DDoS packet senario, XDP >>>>>>> simply passes the packet to Linux as if nothing happened. so it is up >>>>>>> to the XDP program logic, also even if ipfire has the kernel feature >>>>>>> enabled, users still need to attach XDP program to the interface, if >>>>>>> no XDP program is attached to the interface, nothing is in the way to >>>>>>> stop packet flowing to IPfire filtering. >>>>>>> Again this diagram is great to describe the packet path >>>>>>> https://upload.wikimedia.org/wikipedia/commons/3/37/Netfilter-packet-flow.s…, >>>>>>> so the XDP_PASS action from XDP program is to pass packet to Linux as >>>>>>> usual. from the same diagram, you also see "AF_PACKET", right? that is >>>>>>> where tcpdump taking place to capture packet for network >>>>>>> troubleshooting, but tcpdump would not stops all the scenario above >>>>>>> you mentioned working, actually tcpdump is based on classic BPF >>>>>>> technology, the XDP/eBPF is to extended classic BPF technology, it not >>>>>>> only can clone (packet capture), but can also drop, redirect. Please >>>>>>> read more about eBPF/XDP in general if you would like, many online >>>>>>> resources explain better than me :) > Those three are just a few, but >>>>>>> they are commonly used features of IPFire and without them, it would >>>>>>> not be what it is. >>>>>> >>>>>> Thank you. I am very familiar with how Netfilter works, including BPF and XDP. >>>>>> >>>>>> I think your diagram just proves my point when I say that everything is going to bypass the OS. That is the long arrow at the bottom. >>>>>> >>>>> I think you refer to the XDP_TX action to bypass the whole OS, for >>>>> IPFire, this is not recommended >>>> >>>> I think it is safe to assume that this is the core feature of XDP is what most people mean when they refer to it. >>>> >>>>>> At least I am assuming that you are interested in forwarding packets instead of going the “XDP_PASS” route, because for that you don’t need XDP? >>>>>> >>>>> >>>>> For IPFire, XDP_PASS action is recommended because we don't want to >>>>> bypass the OS, we only want to drop DDoS packet at the driver, the >>>>> good packet passes through the OS as usual. >>>> >>>> You did not make at clear at all that your goal is to implement SYNPROXY with BPF. >>>> >>>> You still do not want to bypass the OS, you just want to bypass Netfilter. So let’s maybe try to be more clear with what we are referring to so that we will save many roundtrips. >>>> >>> >>> I assumed everyone knows what I was talking about in my first email >>> with the https://netdevconf.info/0x15/slides/30/Netdev%200x15%20Accelerating%20synpr…, >>> apparently I failed :) >> >> Well, I don’t think it is very efficient to ask people to read through a long presentation to find the thing that you want. You should briefly explain it in your own words and link further references. >> >>>>>> I think you still haven’t explained what your goal is on a lower level. DoS protection is incredibly broad and that does not strictly require XDP. What kind of XDP program are you interested in using? >>>>>> >>>>> use XDP to stop DoS has low overhead, save OS cycles, I have IPFire >>>>> KVM instance, if I run TCP SYN flood the IPFire, the IPFire ssh >>>>> session, WeUI session because sluggish and unresponsive, but with XDP, >>>>> it is almost like nothing happening, IPFire is responsive all the time >>>>> during flood attack. >>>> >>>> Could you describe your test scenario more, please? >>>> >>>> The Linux kernel is already using SYN cookies whenever it cannot keep up with processing all SYN packets that it receives. That can be configured with net.ipv4.tcp_max_syn_backlog which is set to 256 on my system. That should be low enough to trigger the mechanism. >>>> >>> yes, that is for protection for services listening on the Linux host itself. >> >> Ah okay, I wasn’t aware that we were talking about port forwardings. >> >>>> Using a KVM-virtualised machine is probably not the best way to deploy this in production as interrupts are expensive and you are competing for compute power with other machines on the host. But I am sure this is just your test environment. >>>> >>> This is my hobby project so I used KVM, but also I used a virtual >>> machine for the client starting SYN flood, so it is comparable. If I >>> can afford a physical server with 10/25G NIC and a client machine with >>> the same capability, the result will likely be the same. >> >> I highly doubt that as virtual machines are not very good at interrupt handling. >> >>>>> these are the XDP program I am interested in using >>>>> https://elixir.bootlin.com/linux/latest/source/tools/testing/selftests/bpf/… >>>> >>>> Okay, so this does not DoS protect in a strict sense. It just moves the part of SYNPROXY into the driver. I believe this is your primary goal here. >>>> >>> right, it does what SYNPROXY does, but way more efficiently, page 24 >>> of above xdp pdf link shows result. >> >> It does, but it also does less. >> >>> >>>>> https://github.com/NLnetLabs/XDPeriments >>>>> >>>>> I had ported them to xdp-tools >>>>> https://github.com/vincentmli/xdp-tools/tree/vli-xdp-synproxy so >>>>> xdp-tools loader program could attach multiple XDP programs to red0 >>>>> interface to stop various DDoS attack >>>> >>>> Thank you for sharing this. >>>> >>>>> >>>>>>>> >>>>>>>>>> As far as I am aware, IPFire is currently able to handle 25 GBit/sec. on >>>>>>>>>> the right hardware, and SYN flooding attacks are not a major threat to >>>>>>>>>> IPFire users, given that we have historically implemented some fine-tuning >>>>>>>>>> to make such attacks less viable. >>>>>>>>> >>>>>>>>> DDoS attacks to IPFire users do not happen now does not mean it will >>>>>>>>> not happen in the future, SYN flood is just one scenario, so better be >>>>>>>>> prepared than sorry later :) One IPFire user had asked for help >>>>>>>>> https://community.ipfire.org/t/filter-out-ddos-attacks-anyone-can-help-me-p… >>>>>>>> >>>>>>>> Is this only about SYN flooding? >>>>>>>> for layer 4 TCP DDoS, the most common scenario is SYN flooding, ACK flooding, RST flooding. all these flooding can be stopped by SYN cookie that is already built in the Linux TCP host stack, but IPFire is a middle box firewall, the packet destination endpoint is not IPFire, but the host/green network port forwarded by IPFire, so that is where the netfilter SYNPROXY module plays in, I don't see SYNPROXY module being referenced anywhere in IPFire, so even without XDP, I still recommend IPFire provides user option to use SYNPROXY for TCP SYN/ACK/RST flood attack >>>>>>>>> I have studied IPFire, I do not see relevant SYN flooding or DDoS >>>>>>>>> tuning, where is it? netfilter with SYNPROXY module? or the TCP stack >>>>>>>>> syncookie implementation, or suricata ddos rules...etc? keep in mind >>>>>>>>> all these are handled in software, no hardware acceleration. >>>>>>>> >>>>>>>> Yes, IPFire runs in software. We cannot use hardware acceleration because it is designed to pass packets and not to do what we are doing here. >>>>>>> hardware acceleration probably is not the right word here for XDP >>>>>>> because XDP is actually still in the software driver, not inside the >>>>>>> hardware, though there is one hardware vendor that supports running >>>>>>> XDP byte code inside the hardware itself for true hardware >>>>>>> acceleration, but that is not common. again XDP does not interfere >>>>>>> with the IPfire filter except in DDoS scenarios, users can have the >>>>>>> option to drop the packet early in the network driver without >>>>>>> consuming IPfire CPU/memory resource. > >>>>>> >>>>>> But do you have any kind of system out there that is under constant fire and the OS cannot cope? What kind of packet rates or bandwidth are we talking about? >>>>> >>>>> I don't have IPFire in production since I am just starting to know >>>>> about the IPFire project. My day time job is enterprise network >>>>> engineer supporting fortune 500 enterprise customer with our >>>>> enterprise product (BIG-IP) which handles 10G/25G/40G or even 100G >>>>> throughput with FPGA, we often has enterprise customer under DDoS >>>>> attack, and sometime my day time job is to simulate such high >>>>> bandwidth attack in lab to see if enterprise product handles well or >>>>> not. I don't think IPFire can handle such flood attack since I know >>>>> the limitation of netfilter ( with more than 20 years of working with >>>>> Linux networking :)) >>>> >>>> Do you have any figures how much your test environment can handle now compared to a stock IPFire without your changes? >>>> >>> I don't, but I think the xdp pdf link I referred to should answer, if >>> the performance is not significant, the kernel community would not >>> accept such feature. >> >> I don’t think that that is a true statement. Not everything is about performance. >> >> And if something works in a lab with a for-purpose built kernel does not exactly mean that this is true for IPFire. I am sure that in all sorts of tests, there are is no NAT involved which quite likely is always involved in a standard IPFire setup - that is how port forwardings work. Therefore it would be good to have real figures. > > I will see if I can do some testing on real hardware with NAT port forwarding > >> >>>>>>>> IPFire uses SYN cookies by default for all incoming connections. We currently do not use the SYNPROXY module, but that is simply because there has not been any demand for it. If this suits your use-case I would rather implement that than XDP. >>>>>>> home users very unlikely would have this demand because there isn't >>>>>>> much gain for attackers who would use DDoS to attack home users. For >>>>>>> small/medium size businesses, attackers could start DDoS because >>>>>>> business could be impacted and lose profit when under DDoS attack, >>>>>>> many businesses choose cloud DDoS providers if they could not afford >>>>>>> DDoS protection devices. XDP DDoS protection on IPFire provides DDoS >>>>>>> protection on inexpensive commodity hardware. There are already a lot >>>>>>> of open source XDP programs out there, including XDP SYNCookie from >>>>>>> Linux kernel source, I have ported it to xdp-tool repo and ported >>>>>>> xdp-tool to IPfire. >>>>>> >>>>>> Maybe share this code on the list here so that people understand better what you are looking for. >>>>>> >>>>> see above link I shared >>>>> >>>>>> I am still not sure what kind of changes you are asking us to make. >>>>>> >>>>> actually the asking is minimum, turn on the kernel config feature >>>>> for eBPF for networking >>>> >>>> You say as a minimum, but you have not asked this before. Usually we handle this in the way that people send an email with a description of their feature and if there is generally an option to include this in the distribution. If agreed, then you can work on some code and send it to this list. >>>> >>> yep, should have asked this in the beginning. >> >> Yes, you should also have sent your code that you have been working on and sent a link to your YouTube channel (https://www.youtube.com/@BPFireOS) where you explain your feature in detail. That would have helped a lot to understand where you were actually going with this instead of us trying to fill in the blanks. >> >>>>> CONFIG_BPF_SYSCALL=y >>>>> CONFIG_DEBUG_INFO=y >>>>> CONFIG_DEBUG_INFO_BTF=y >>>>> CONFIG_DEBUG_INFO_DWARF4=y >>>>> CONFIG_BPF_UNPRIV_DEFAULT_OFF=y >>>> >>>> We don’t need any of those debugging symbols in production. They are actually really large and will make the kernel slower. >>>> >>> the debug symbols is required during build time for kernel image, but >>> can be stripped after BTF is generated for kernel image >> >> How much extra build time does generating the debug information add? >> > > I did not measure with and without debug info, my impression is it is > not significantly long that I would have noticed, I can definitely > measure. > >>>> I have just posted a suggestion to the list as we are going to ship a fresh kernel with the next Core Update: >>>> >>>> https://lists.ipfire.org/hyperkitty/list/development@lists.ipfire.org/threa… >>>> >>>>> I have a working fork of IPFire >>>>> https://github.com/vincentmli/BPFire/tree/bpfire, the discussion here >>>>> is I want to share that great technology with the IPFire community and >>>>> contribute that to IPFire. >>>> >>>> I was entirely unaware that you have written any code here. And I was also entirely unaware what your goal was. Thankfully we are on the same page now. >>>> >>>> Since I didn’t know that you have written some code, I implemented a classic SYN proxy using Netfilter this afternoon: >>>> >>>> https://lists.ipfire.org/hyperkitty/list/development@lists.ipfire.org/threa… >>>> >>>> I believe that this does what you want to do, although it does not use BPF/XDP. However, it has the advantage that it can be enabled on a per rule basis and does not have to be globally enabled for all incoming connection to that host. How could BPF/XDP be integrated into this without losing that functionality? >>>> >>> enabled kernel config feature for XDP does not mean all incoming >>> traffic will be processed by XDP, the incoming traffic will only be >>> processed if there is XDP program bytecode/machine code attached to >>> the incoming network interface, so it is up to what individual XDP >>> program does >>> >>> for example XDP program below, it does nothing, simply pass every >>> packet/connection to the netfilter/OS >>> >>> SEC("xdp") >>> int xdp_pass(struct xdp_md *ctx) >>> { >>> return XDP_PASS; >>> } >>> >>> another XDP program example to drop packet to port 5555 >>> >>> SEC("xdp") >>> int xdp_drop(struct xdp_md *ctx) >>> { >>> ... code to parse the raw packet header.. >>> if tcp->dest == 5555 >>> return XDP_DROP; >>> >>> } >>> >>> so you could attach none or many XDP programs to the network >>> interface, the XDP program developers have to write the XDP program >>> based on what users want. >>> >>> The XDP synproxy program is more complicated than the above example, >>> but it also only does filtering on a per port basis and configurable >>> by user. >>> >>> for example snippet of the code: >>> >>> /* Pass to upper stack if port requires no syncookie handling */ >>> if (!check_port_allowed(bpf_ntohs(hdr->tcp->dest))) >>> return XDP_PASS; >>> >>> so for example if user only want port 80 syn flood protected by the >>> XDP syncookie program , user can add port 80 to the ebpf map that >>> function check_port_allowed looks up ( in my fork I already added >>> IPFire UI option for user to do that), for all other ports, XDP >>> program does nothing about it, simply pass it to netfilter/OS. >> >> Well, this does answer my question then… This solution cannot be enabled individually per rule, but only globally for all traffic that hits the firewall on a certain port. That is not exactly selling this. >> > I am not sure I understand what you mean by per rule, you mean iptable > rules for iptables SYNPROXY module or just iptable rules. for XDP > SYNPROXY to work, it also does require setup iptable rule for each > port that requires XDP SYNPROXY. In the web UI, you can define firewall rules. Those. They might generate more than one iptables rule, but generally that is what I am interested in. You can combine SYNPROXY with other features that we have like country filtering and so on. So there are huge benefits there. The XDP approach that you have been taking seems to take over the entire port no matter where the packet is coming from which makes this solution less flexible. -Michael >> -Michael >> >>> >>>> Best, >>>> -Michael >>>> >>>>> >>>>> Vincent >>>>> >>>>> >>>>>> -Michael >>>>>> >>>>>>>> >>>>>>>> -Michael >>>>>>>> >>>>>>>>> Why not give IPFire users the options when the options already exist >>>>>>>>> in the IPFire kernel? >>>>>>>>> >>>>>>>>>> >>>>>>>>>> Therefore, I - personally - neither see the necessity nor benefit of pursuing >>>>>>>>>> this proposal at this time. >>>>>>>>>> >>>>>>>>>> Thanks, and best regards, >>>>>>>>>> Peter Müller >>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>>> I don't understand what the difference is between XDP_PASS and XDP_TX but I would expect that nothing should be allowed to bypass the netfilter section unless it is being dropped or rejected already by the XDP process. >>>>>>>>>>>> >>>>>>>>>>> >>>>>>>>>>> XDP_PASS is to pass the packet to netfilter/TCP stack as usual after >>>>>>>>>>> XDP program packet processing, XDP_TX is to redirect the packet back >>>>>>>>>>> out through the same network interface after XDP program packet >>>>>>>>>>> processing. >>>>>>>>>>> >>>>>>>>>>>> Regards, >>>>>>>>>>>> >>>>>>>>>>>> Adolf. >>>>>>>>>>>> >>>>>>>>>>>> On 09/04/2024 19:36, Vincent Li wrote: >>>>>>>>>>>>> Hi, >>>>>>>>>>>>> >>>>>>>>>>>>> I have been working on enabling eBPF XDP/TC kernel feature for IPFire, >>>>>>>>>>>>> please refer to >>>>>>>>>>>>> https://upload.wikimedia.org/wikipedia/commons/3/37/Netfilter-packet-flow.s… >>>>>>>>>>>>> for where XDP fit in Linux network datapath, XDP will not interfere >>>>>>>>>>>>> with existing IPFire firewall rules. XDP is especially good at DDoS >>>>>>>>>>>>> packet filtering at high speed, see >>>>>>>>>>>>> https://netdevconf.info/0x15/slides/30/Netdev%200x15%20Accelerating%20synpr… >>>>>>>>>>>>> >>>>>>>>>>>>> I think we only need to enable XDP/TC network filtering capability >>>>>>>>>>>>> without eBPF tracing capability which some users are concerned about >>>>>>>>>>>>> potential host security information leaks. >>>>>>>>>>>>> >>>>>>>>>>>>> Please let me know what you think, thanks! >>>>>>>>>>>>> >>>>>>>>>>>>> Vincent

1 0

[RFC PATCH] kernel: Enable XDP
by Michael Tremer 24 Apr '24

24 Apr '24

https://lists.ipfire.org/hyperkitty/list/development@lists.ipfire.org/threa… Signed-off-by: Michael Tremer <michael.tremer(a)ipfire.org> --- config/kernel/kernel.config.aarch64-ipfire | 11 ++++++++++- config/kernel/kernel.config.riscv64-ipfire | 13 ++++++++++++- config/kernel/kernel.config.x86_64-ipfire | 11 ++++++++++- 3 files changed, 32 insertions(+), 3 deletions(-) diff --git a/config/kernel/kernel.config.aarch64-ipfire b/config/kernel/kernel.config.aarch64-ipfire index b85d7add9..3d8282ac6 100644 --- a/config/kernel/kernel.config.aarch64-ipfire +++ b/config/kernel/kernel.config.aarch64-ipfire @@ -93,9 +93,12 @@ CONFIG_ARCH_WANT_DEFAULT_BPF_JIT=y # # BPF subsystem # -# CONFIG_BPF_SYSCALL is not set +CONFIG_BPF_SYSCALL=y CONFIG_BPF_JIT=y +CONFIG_BPF_JIT_ALWAYS_ON=y CONFIG_BPF_JIT_DEFAULT_ON=y +CONFIG_BPF_UNPRIV_DEFAULT_OFF=y +# CONFIG_BPF_PRELOAD is not set # end of BPF subsystem CONFIG_PREEMPT_BUILD=y @@ -136,6 +139,7 @@ CONFIG_TREE_SRCU=y CONFIG_TASKS_RCU_GENERIC=y CONFIG_TASKS_RCU=y CONFIG_TASKS_RUDE_RCU=y +CONFIG_TASKS_TRACE_RCU=y CONFIG_RCU_STALL_COMMON=y CONFIG_RCU_NEED_SEGCBLIST=y # end of RCU Subsystem @@ -179,6 +183,7 @@ CONFIG_PROC_PID_CPUSET=y CONFIG_CGROUP_DEVICE=y CONFIG_CGROUP_CPUACCT=y CONFIG_CGROUP_PERF=y +# CONFIG_CGROUP_BPF is not set # CONFIG_CGROUP_MISC is not set # CONFIG_CGROUP_DEBUG is not set CONFIG_SOCK_CGROUP_DATA=y @@ -1036,6 +1041,8 @@ CONFIG_XFRM_ESP=m CONFIG_XFRM_IPCOMP=m CONFIG_NET_KEY=m CONFIG_NET_KEY_MIGRATE=y +CONFIG_XDP_SOCKETS=y +CONFIG_XDP_SOCKETS_DIAG=m CONFIG_NET_HANDSHAKE=y CONFIG_INET=y CONFIG_IP_MULTICAST=y @@ -1144,6 +1151,7 @@ CONFIG_NETFILTER_SKIP_EGRESS=y CONFIG_NETFILTER_NETLINK=m CONFIG_NETFILTER_FAMILY_BRIDGE=y CONFIG_NETFILTER_FAMILY_ARP=y +CONFIG_NETFILTER_BPF_LINK=y # CONFIG_NETFILTER_NETLINK_HOOK is not set CONFIG_NETFILTER_NETLINK_ACCT=m CONFIG_NETFILTER_NETLINK_QUEUE=m @@ -6466,6 +6474,7 @@ CONFIG_HID_ZYDACRON=m # # HID-BPF support # +# CONFIG_HID_BPF is not set # end of HID-BPF support # diff --git a/config/kernel/kernel.config.riscv64-ipfire b/config/kernel/kernel.config.riscv64-ipfire index 2bd39e7de..fd2172a25 100644 --- a/config/kernel/kernel.config.riscv64-ipfire +++ b/config/kernel/kernel.config.riscv64-ipfire @@ -103,8 +103,13 @@ CONFIG_HAVE_EBPF_JIT=y # # BPF subsystem # -# CONFIG_BPF_SYSCALL is not set +CONFIG_BPF_SYSCALL=y CONFIG_BPF_JIT=y +CONFIG_BPF_JIT_ALWAYS_ON=y +CONFIG_BPF_JIT_DEFAULT_ON=y +CONFIG_BPF_UNPRIV_DEFAULT_OFF=y +# CONFIG_BPF_PRELOAD is not set +# CONFIG_BPF_LSM is not set # end of BPF subsystem CONFIG_PREEMPT_VOLUNTARY_BUILD=y @@ -139,6 +144,7 @@ CONFIG_TREE_RCU=y CONFIG_TREE_SRCU=y CONFIG_TASKS_RCU_GENERIC=y CONFIG_TASKS_RUDE_RCU=y +CONFIG_TASKS_TRACE_RCU=y CONFIG_RCU_STALL_COMMON=y CONFIG_RCU_NEED_SEGCBLIST=y # end of RCU Subsystem @@ -180,6 +186,7 @@ CONFIG_PROC_PID_CPUSET=y CONFIG_CGROUP_DEVICE=y CONFIG_CGROUP_CPUACCT=y CONFIG_CGROUP_PERF=y +# CONFIG_CGROUP_BPF is not set # CONFIG_CGROUP_MISC is not set # CONFIG_CGROUP_DEBUG is not set CONFIG_SOCK_CGROUP_DATA=y @@ -826,6 +833,8 @@ CONFIG_XFRM_ESP=m CONFIG_XFRM_IPCOMP=m CONFIG_NET_KEY=m CONFIG_NET_KEY_MIGRATE=y +CONFIG_XDP_SOCKETS=y +CONFIG_XDP_SOCKETS_DIAG=m CONFIG_NET_HANDSHAKE=y CONFIG_INET=y CONFIG_IP_MULTICAST=y @@ -934,6 +943,7 @@ CONFIG_NETFILTER_SKIP_EGRESS=y CONFIG_NETFILTER_NETLINK=y CONFIG_NETFILTER_FAMILY_BRIDGE=y CONFIG_NETFILTER_FAMILY_ARP=y +CONFIG_NETFILTER_BPF_LINK=y # CONFIG_NETFILTER_NETLINK_HOOK is not set CONFIG_NETFILTER_NETLINK_ACCT=m CONFIG_NETFILTER_NETLINK_QUEUE=m @@ -7383,6 +7393,7 @@ CONFIG_BRANCH_PROFILE_NONE=y # CONFIG_PROFILE_ANNOTATED_BRANCHES is not set # CONFIG_BLK_DEV_IO_TRACE is not set CONFIG_UPROBE_EVENTS=y +CONFIG_BPF_EVENTS=y CONFIG_DYNAMIC_EVENTS=y CONFIG_PROBE_EVENTS=y CONFIG_FTRACE_MCOUNT_RECORD=y diff --git a/config/kernel/kernel.config.x86_64-ipfire b/config/kernel/kernel.config.x86_64-ipfire index 8ce66cb4f..ec3bca2e7 100644 --- a/config/kernel/kernel.config.x86_64-ipfire +++ b/config/kernel/kernel.config.x86_64-ipfire @@ -113,9 +113,12 @@ CONFIG_ARCH_WANT_DEFAULT_BPF_JIT=y # # BPF subsystem # -# CONFIG_BPF_SYSCALL is not set +CONFIG_BPF_SYSCALL=y CONFIG_BPF_JIT=y +CONFIG_BPF_JIT_ALWAYS_ON=y CONFIG_BPF_JIT_DEFAULT_ON=y +CONFIG_BPF_UNPRIV_DEFAULT_OFF=y +# CONFIG_BPF_PRELOAD is not set # end of BPF subsystem CONFIG_PREEMPT_BUILD=y @@ -156,6 +159,7 @@ CONFIG_TREE_SRCU=y CONFIG_TASKS_RCU_GENERIC=y CONFIG_TASKS_RCU=y CONFIG_TASKS_RUDE_RCU=y +CONFIG_TASKS_TRACE_RCU=y CONFIG_RCU_STALL_COMMON=y CONFIG_RCU_NEED_SEGCBLIST=y # end of RCU Subsystem @@ -199,6 +203,7 @@ CONFIG_PROC_PID_CPUSET=y CONFIG_CGROUP_DEVICE=y CONFIG_CGROUP_CPUACCT=y CONFIG_CGROUP_PERF=y +# CONFIG_CGROUP_BPF is not set # CONFIG_CGROUP_MISC is not set # CONFIG_CGROUP_DEBUG is not set CONFIG_SOCK_CGROUP_DATA=y @@ -1099,6 +1104,8 @@ CONFIG_XFRM_ESP=m CONFIG_XFRM_IPCOMP=m CONFIG_NET_KEY=m CONFIG_NET_KEY_MIGRATE=y +CONFIG_XDP_SOCKETS=y +CONFIG_XDP_SOCKETS_DIAG=m CONFIG_NET_HANDSHAKE=y CONFIG_INET=y CONFIG_IP_MULTICAST=y @@ -1207,6 +1214,7 @@ CONFIG_NETFILTER_SKIP_EGRESS=y CONFIG_NETFILTER_NETLINK=y CONFIG_NETFILTER_FAMILY_BRIDGE=y CONFIG_NETFILTER_FAMILY_ARP=y +CONFIG_NETFILTER_BPF_LINK=y # CONFIG_NETFILTER_NETLINK_HOOK is not set CONFIG_NETFILTER_NETLINK_ACCT=m CONFIG_NETFILTER_NETLINK_QUEUE=m @@ -5861,6 +5869,7 @@ CONFIG_HID_SENSOR_HUB=m # # HID-BPF support # +# CONFIG_HID_BPF is not set # end of HID-BPF support # -- 2.39.2

3 5

Enable eBPF XDP/TC kernel feature for IPFire
by Vincent Li 24 Apr '24

24 Apr '24

Hi, I have been working on enabling eBPF XDP/TC kernel feature for IPFire, please refer to https://upload.wikimedia.org/wikipedia/commons/3/37/Netfilter-packet-flow.s… for where XDP fit in Linux network datapath, XDP will not interfere with existing IPFire firewall rules. XDP is especially good at DDoS packet filtering at high speed, see https://netdevconf.info/0x15/slides/30/Netdev%200x15%20Accelerating%20synpr… I think we only need to enable XDP/TC network filtering capability without eBPF tracing capability which some users are concerned about potential host security information leaks. Please let me know what you think, thanks! Vincent

4 12

[PATCH 1/2] libhtp: Update to 0.5.48
by Michael Tremer 24 Apr '24

24 Apr '24

https://github.com/OISF/libhtp/releases/tag/0.5.48 Signed-off-by: Michael Tremer <michael.tremer(a)ipfire.org> --- lfs/libhtp | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/lfs/libhtp b/lfs/libhtp index cbd4cb093..d3b56dcb2 100644 --- a/lfs/libhtp +++ b/lfs/libhtp @@ -24,7 +24,7 @@ include Config -VER = 0.5.47 +VER = 0.5.48 THISAPP = libhtp-$(VER) DL_FILE = $(THISAPP).tar.gz @@ -40,7 +40,7 @@ objects = $(DL_FILE) $(DL_FILE) = $(DL_FROM)/$(DL_FILE) -$(DL_FILE)_BLAKE2 = a2a126052ea528b84dfffd7926a80e633f649968bf098e3ff0f49d441821723a4d3f78f0f15d530e9ce9e393c6173aceced8c0d863afecbe65c736e4ab874a03 +$(DL_FILE)_BLAKE2 = 56eeef3524ffeac593c251846196d09e0ccfacd9aadd03b35061fd1fe6a245d1374c338581b4f7bee67255797740f4e282344fc10bf3d0c0fdf824f159380053 install : $(TARGET) -- 2.39.2

1 1

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

Development April 2024