This reverts commit e0be9eab47.
This change is now producing problems on IPv6-enabled systems as it will
deny access to any website that is IPv6-enabled as well, even if the
client connected using IPv4.
I have tested if squid is now running on fine on systems where IPv6 is
disabled and can confirm that its running just fine.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Acked-by: Peter Müller <peter.mueller@ipfire.org>
The function tries to figure out which networks are connected locally,
but VPN tunnels that use 0.0.0.0 and GRE/VTI interfaces will be
considered local and the proxy is being disabled for everyone.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
- This line is no longer needed with the help button at the top of each WUI page that was
implemented by @Leo
- Sorry for long time for me to get around to finalising this patch but it is done now.
Fixes: bug#12701
Tested-by: Adolf Belka <adolf.belka@ipfire.org>
Signed-off-by: Adolf Belka <adolf.belka@ipfire.org>
Bumping across one of our scripts with very long trailing whitespaces, I
thought it might be a good idea to clean these up. Doing so, some
missing or inconsistent licence headers were fixed.
There is no need in shipping all these files en bloc, as their
functionality won't change.
Signed-off-by: Peter Müller <peter.mueller@ipfire.org>
There is no sense to display this to anybody, and we do not reveal
version information anywhere else on purpose. The IT staff knows which
version of IPFire they are running (hopefully the latest), and it's
none of the rest of the world's business.
Fixes: #12665 (in some way)
Signed-off-by: Peter Müller <peter.mueller@ipfire.org>
Acked-by: Michael Tremer <michael.tremer@ipfire.org>
Signed-off-by: Arne Fitzenreiter <arne_f@ipfire.org>
This patch adds two new features to IPFire's web proxy:
(a) Proactive Fast Flux detection
FQDNs are resolved to their IP addresses, which are then resolved to
corresponding Autonomous System Numbers using IPFire's location
database. Most destinations will scatter across a very low number of
ASNs (not to be confused with IP addresses!). FQDNs hosted on Fast
Flux setups have a significantly higher ASN diversity (5 is usually
a good threshold), so they can be proactively detected.
(b) Detection for selectively announced destinations
Especially in targeted operations, miscreants host FQDNs for
exfiltrating data or malware distributions on ASNs not announced
globally, but only to the intended victim or it's upstream ISPs.
That way, security researchers located in other parts of the
internet have no insights into these attacks, hence not being able
to publish listings or send take down notices for the domains used.
While RPKI made this attack harder, it can still be observed every
now and then.
This feature also protects against accessing FQDNs resolving to IP
addresses not being globally routeable, hence providing a trivial
mitigation for so-called "rebound attacks" - which we cannot filter
at DNS level currently.
The second version of this patch consumes the user-defined whitelist for
the URL filter (if present and populated) for the ASNBL helper as well,
to make exceptions for funny destinations such as fedoraproject.org
possible. In addition, the ASNBL helper's sanity tests no longer include
publicly routable IP addresses, so failures on location01 cannot brick
IPFire installations in the field.
Thanks to Michael Tremer and Adolf Belka for these suggestions.
Signed-off-by: Peter Müller <peter.mueller@ipfire.org>
Reviewed-by: Michael Tremer <michael.tremer@ipfire.org>
Signed-off-by: Arne Fitzenreiter <arne_f@ipfire.org>
While hiding version information does not come with any _actual_
security improvements, it is generally a good thing to do so by default:
Attackers will still be able to reasonably guess or enumerate the
software version running, but need to conduct additional effort to do
so, hence more likely raising alerts and drawing attention on their
operation.
In addition, we suppress version details somewhere else in IPFire 2.x by
default, too (e. g. Unbound and Apache), so we can justify this patch by
aiming to stay consistent, I guess. :-)
Signed-off-by: Peter Müller <peter.mueller@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
While maintaining privacy when accessing web sites probably has never
been more important than it is today, faking Referer and User-Agent
headers is both obsolete and counterproductive:
(a) Most web sites require HTTPS, thwarting manipulation attempts to
HTTP headers in transit. Given todays' internet landscape, faking
these headers is unlikely to work for the vast majority of web
sites.
(b) It is trivial to detect faked HTTP User-Agent headers by obtaining
corresponding browser information via JavaScript. Any difference
most likely indicates (trivial) header manipulation attempts, hence
rendering this feature useless if browsers do not behave in the same
manner, which we cannot control on IPFire.
(c) Especially static Referer headers make users stick out like a sore
thumb, as nobody else in the world is likely to have the same
Referer set _all the time_.
Modern browsers attempt to strip sensitive information from Referer
headers, or ditch them completely, particularly to 3rd party sites.
Given the state of the web ecosystem as we know it today, enforcing
privacy in a centralised manner does not even come close to being
sufficient. Without gaining control over users' browsers, their
settings, and their infrastructure (such as setting up terminal
environments for accessing the web, preventing hardware
fingerprinting), a centralised attempt will at best fail, if not making
things worse, as highlighted in (c).
Therefore, removing these features from the Squid GUI is the least worse
option we have. We should not give our users a false sense of privacy.
Signed-off-by: Peter Müller <peter.mueller@ipfire.org>
Reviewed-by: Bernhard Bitsch <bbitsch@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
This disables the theme support and makes it impossible to use any other
themes than the ipfire default theme.
The only intention of this patch is to hardcode the theme to ipfire.
To change any cgi we have is an ugly way, but the only way to do this
fast. The colour handling needs certainly to be improved as well, but
this will and should be done in other patches.
Signed-off-by: Jonatan Schlag <jonatan.schlag@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
This reverts commit dc637f087f.
Rationale: "authenticate_ip_ttl" can be safely used as it does not
introduces an authentication bypass, but saves relationships between
successfully authenticated users and their IP addresses.
"max_user_ip" depends on such an authentication cache, so credential
sharing between several IPs (on purpose or by chance) can be detected
properly. This is useful in case of crompromised machines and/or
attackers in internal networks having stolen proxy authentication
credentials.
Quoted from squid.conf.documented or man 5 squid.conf:
> acl aclname max_user_ip [-s] number
> # This will be matched when the user attempts to log in from more
> # than <number> different ip addresses. The authenticate_ip_ttl
> # parameter controls the timeout on the ip entries. [fast]
> # If -s is specified the limit is strict, denying browsing
> # from any further IP addresses until the ttl has expired. Without
> # -s Squid will just annoy the user by "randomly" denying requests.
> # (the counter is reset each time the limit is reached and a
> # request is denied)
> # NOTE: in acceleration mode or where there is mesh of child proxies,
> # clients may appear to come from multiple addresses if they are
> # going through proxy farms, so a limit of 1 may cause user problems.
Fixes: #11994
Cc: Michael Tremer <michael.tremer@ipfire.org>
Signed-off-by: Peter Müller <peter.mueller@ipfire.org>
Signed-off-by: Arne Fitzenreiter <arne_f@ipfire.org>
With this commit suricata reads the HTTP port declarations from a newly
introduced external file
(/var/ipfire/suricata/suricata-http-ports.yaml).
This file dynamically will be generated. HTTP ports always are the
default port "80" and "81" for update Accelerator and HTTP access to the
WUI. In case the Web-proxy is used, the configured proxy port and/or Transparent
Proxy port also will be declared as a HTTP port and written to that file.
In case one of the proxy ports will be changed, the HTTP port file will
be re-generated and suricate restarted if launched. Also if an old
backup with snort will be restored the convert script handles the
generation of the HTTP ports file.
Finally the suricata-generate-http-ports-file as a tiny script which
simply generates the http ports file and needs to be launched during the
installation of a core update. (The script will no be required
anymore, so it could be deleted afterwards.)
Fixes#12308.
Signed-off-by: Stefan Schantl <stefan.schantl@ipfire.org>
Signed-off-by: Arne Fitzenreiter <arne_f@ipfire.org>
The former used default Crypt algorithmus only supports passwords up to
eight signs wheater MD5 does not have any limitation here.
Fixes 12290.
Signed-off-by: Stefan Schantl <stefan.schantl@ipfire.org>
Reviewed-by: Michael Tremer <michael.tremer@ipfire.org>
Signed-off-by: Arne Fitzenreiter <arne_f@ipfire.org>
This patch fixes the behavior in 11696 and adds IPSEC and OpenVPN n2n subnets to wpad.dat so they don't pass through the proxy.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
This patch adds a notice with a link to the Wiki-page https://wiki.ipfire.org/configuration/network/proxy/extend/wpad to the new WebGUI-Setion to make the user aware of the fact, that WPAD will only work correctly if he makes further adjustments:
- Add DHCP-Options for WPAD via DHCP
- Add HOST-Entries to DNS and Apache-vhost or haproxy-frontend/backend or firewall-redirect for WPAD via DNS
These additional options depend on the users environment and can not be shipped by default as they might break the users setups.
Note: The translations are only done for "en" and "de" yet!
Signed-off-by: Alexander Koch <ipfire@starkstromkonsument.de>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
This patch adds the missing Web-GUI for the WPAD-Exceptions to proxy.cgi
Note: The translations are only done for "en" and "de" yet!
Signed-off-by: Alexander Koch <ipfire@starkstromkonsument.de>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
This patch extends the script /srv/web/ipfire/cgi-bin/proxy.cgi by additional code for reading exceptions for URL's and IP's/Subnets from two new files:
- /var/ipfire/proxy/advanced/acls/dst_noproxy_url.acl
- /var/ipfire/proxy/advanced/acls/dst_noproxy_ip.acl
as described in: https://wiki.ipfire.org/configuration/network/proxy/extend/add_distri
These can be used to define additional URL's, IP's and Subnets that should be retrieved "DIRECT" and not via the proxy. The files have to be created by the user, as the WPAD-Feature is not enabled by default anyway. If the files are not present or their size is 0, nothing is done. I'll revise the wiki-page, after the patch is merged and the core update is released.
Signed-off-by: Alexander Koch <ipfire@starkstromkonsument.de>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
This is potentially dangerous to set larger than zero.
Authentication is perfomed on basis of IP addresses which is
not a good idea at all.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
This is the authentication againt NT 4.0 style domain controllers.
squid has dropped support for this in the 4.5 release and nobody
should be using these old domain controllers any more.
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
I added a function to determine the number of cores.
Now the number of squid processes will be equal to the number of logical cores.
Further I removed the possibility of changing the number
of squid processes in the proxy.cgi
Signed-off-by: Daniel Weismüller <daniel.weismueller@ipfire.org>
Signed-off-by: root <root@ipfire.test>
As suggested by Oliver "giller" Fieker <oli@new-lan.de>
in bug 10592 I added the functionality to use the squid as ram-only cache.
Further it defines the maximum_object_size_in_memory
as 2% of the in the webif defined "Memory cache size".
The maximum_object_size_in_memory should have a useful
size of the defined memory cache and I don't want to
create another variable which muste be fulled in by the user.
Signed-off-by: Daniel Weismüller <daniel.weismueller@ipfire.org>
Suggested-by: Oliver "giller" Fieker <oli@new-lan.de>
Suggested-by: Kim Wölfel <xaver4all@gmx.de>
Acked-by: Michael Tremer <michael.tremer@ipfire.org>
Cc: Stefan Schantl <stefan.schantl@ipfire.org>
Signed-off-by: Daniel Weismüller <daniel.weismueller@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Another clickable link for 'proxy.cgi', this time for 'Cache Manager Interface' - this one opens in a new window.
And: This time - hopefully - with correct '_blank'-attribute (deleted the backslashes) - based on current 'next'.
Plus: Deleted some "blind" tabs - found by chance.
Signed-off-by: Matthias Fischer <matthias.fischer@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Added clickable links for 'URL filter' and 'Update accelerator' for faster access,
this time without the need to alter the language-files.
Best,
Matthias
Signed-off-by: Matthias Fischer <matthias.fischer@ipfire.org>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
The correct case for "kilobit" is "kilobit", not "kiloBit".
And the same applies for Mbit, Gbit etc.
Reference is https://en.wikipedia.org/wiki/Kilobit
This commit changes the texts used in the web UI, so
that it correctly displays as "bit", "kbit", "Mbit" etc.
This fixes bugzilla item 10918.
Signed-off-by: Alf Høgemark <alf@i100.no>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Handcrafted requests with shell commands could be sent to these
CGI files and gain shell access as unprivileged user.
References: #11087
Reported-by: Yann Cam <yann.cam@gmail.com>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Mark required input fields with a star as nowadays this is
the de-facto default. Before, it was the other way around and
optional fields were marked.
Signed-off-by: Lars Schumacher <larsen007@web.de>
Signed-off-by: Michael Tremer <michael.tremer@ipfire.org>
Some clients may not support NTLMv2. Basic authentication can
now be activated. This is dangerous as it sends the credentials
in cleartext to the proxy server.
