X-Git-Url: https://git.exim.org/exim.git/blobdiff_plain/7c261cc6918d1bdb019a213fc986b668cbbe456e..fca41d5a245023376c7d7716a3f84abc2aaa4b8e:/doc/doc-txt/experimental-spec.txt diff --git a/doc/doc-txt/experimental-spec.txt b/doc/doc-txt/experimental-spec.txt index b33612f43..80e970cc1 100644 --- a/doc/doc-txt/experimental-spec.txt +++ b/doc/doc-txt/experimental-spec.txt @@ -6,114 +6,6 @@ about experimental features, all of which are unstable and liable to incompatible change. -PRDR support --------------------------------------------------------------- - -Per-Recipient Data Reponse is an SMTP extension proposed by Eric Hall -in a (now-expired) IETF draft from 2007. It's not hit mainstream -use, but has apparently been implemented in the META1 MTA. - -There is mention at http://mail.aegee.org/intern/sendmail.html -of a patch to sendmail "to make it PRDR capable". - - ref: http://www.eric-a-hall.com/specs/draft-hall-prdr-00.txt - -If Exim is built with EXPERIMENTAL_PRDR there is a new config -boolean "prdr_enable" which controls whether PRDR is advertised -as part of an EHLO response, a new "acl_data_smtp_prdr" ACL -(called for each recipient, after data arrives but before the -data ACL), and a new smtp transport option "hosts_try_prdr". - -PRDR may be used to support per-user content filtering. Without it -one must defer any recipient after the first that has a different -content-filter configuration. With PRDR, the RCPT-time check -for this can be disabled when the MAIL-time $smtp_command included -"PRDR". Any required difference in behaviour of the main DATA-time -ACL should however depend on the PRDR-time ACL having run, as Exim -will avoid doing so in some situations (eg. single-recipient mails). - - - -OCSP Stapling support --------------------------------------------------------------- - -X.509 PKI certificates expire and can be revoked; to handle this, the -clients need some way to determine if a particular certificate, from a -particular Certificate Authority (CA), is still valid. There are three -main ways to do so. - -The simplest way is to serve up a Certificate Revocation List (CRL) with -an ordinary web-server, regenerating the CRL before it expires. The -downside is that clients have to periodically re-download a potentially -huge file from every certificate authority it knows of. - -The way with most moving parts at query time is Online Certificate -Status Protocol (OCSP), where the client verifies the certificate -against an OCSP server run by the CA. This lets the CA track all -usage of the certs. This requires running software with access to the -private key of the CA, to sign the responses to the OCSP queries. OCSP -is based on HTTP and can be proxied accordingly. - -The only widespread OCSP server implementation (known to this writer) -comes as part of OpenSSL and aborts on an invalid request, such as -connecting to the port and then disconnecting. This requires -re-entering the passphrase each time some random client does this. - -The third way is OCSP Stapling; in this, the server using a certificate -issued by the CA periodically requests an OCSP proof of validity from -the OCSP server, then serves it up inline as part of the TLS -negotiation. This approach adds no extra round trips, does not let the -CA track users, scales well with number of certs issued by the CA and is -resilient to temporary OCSP server failures, as long as the server -starts retrying to fetch an OCSP proof some time before its current -proof expires. The downside is that it requires server support. - -If Exim is built with EXPERIMENTAL_OCSP and it was built with OpenSSL, -then it gains a new global option: "tls_ocsp_file". - -The file specified therein is expected to be in DER format, and contain -an OCSP proof. Exim will serve it as part of the TLS handshake. This -option will be re-expanded for SNI, if the tls_certificate option -contains $tls_sni, as per other TLS options. - -Exim does not at this time implement any support for fetching a new OCSP -proof. The burden is on the administrator to handle this, outside of -Exim. The file specified should be replaced atomically, so that the -contents are always valid. Exim will expand the "tls_ocsp_file" option -on each connection, so a new file will be handled transparently on the -next connection. - -Exim will check for a valid next update timestamp in the OCSP proof; -if not present, or if the proof has expired, it will be ignored. - -Also, given EXPERIMENTAL_OCSP and OpenSSL, the smtp transport gains -a "hosts_require_ocsp" option; a host-list for which an OCSP Stapling -is requested and required for the connection to proceed. The host(s) -should also be in "hosts_require_tls", and "tls_verify_certificates" -configured for the transport. - -For the client to be able to verify the stapled OCSP the server must -also supply, in its stapled information, any intermediate -certificates for the chain leading to the OCSP proof from the signer -of the server certificate. There may be zero or one such. These -intermediate certificates should be added to the server OCSP stapling -file (named by tls_ocsp_file). - -At this point in time, we're gathering feedback on use, to determine if -it's worth adding complexity to the Exim daemon to periodically re-fetch -OCSP files and somehow handling multiple files. - - A helper script "ocsp_fetch.pl" for fetching a proof from a CA - OCSP server is supplied. The server URL may be included in the - server certificate, if the CA is helpful. - - One fail mode seen was the OCSP Signer cert expiring before the end - of vailidity of the OCSP proof. The checking done by Exim/OpenSSL - noted this as invalid overall, but the re-fetch script did not. - - - - Brightmail AntiSpam (BMI) suppport -------------------------------------------------------------- @@ -452,15 +344,21 @@ which the spf condition should succeed. Valid strings are: This means the queried domain has published a SPF record, but wants to allow outside servers to send mail under its domain as well. - o err_perm This indicates a syntax error in the SPF - record of the queried domain. This should be - treated like "none". - o err_temp This indicates a temporary error during all + This should be treated like "none". + o permerror This indicates a syntax error in the SPF + record of the queried domain. You may deny + messages when this occurs. (Changed in 4.83) + o temperror This indicates a temporary error during all processing, including Exim's SPF processing. You may defer messages when this occurs. + (Changed in 4.83) + o err_temp Same as permerror, deprecated in 4.83, will be + removed in a future release. + o err_perm Same as temperror, deprecated in 4.83, will be + removed in a future release. You can prefix each string with an exclamation mark to invert -is meaning, for example "!fail" will match all results but +its meaning, for example "!fail" will match all results but "fail". The string list is evaluated left-to-right, in a short-circuit fashion. When a string matches the outcome of the SPF check, the condition succeeds. If none of the listed @@ -510,8 +408,8 @@ variables. $spf_result This contains the outcome of the SPF check in string form, - one of pass, fail, softfail, none, neutral, err_perm or - err_temp. + one of pass, fail, softfail, none, neutral, permerror or + temperror. $spf_smtp_comment This contains a string that can be used in a SMTP response @@ -773,7 +671,7 @@ fails. Of course, you can also use any other lookup method that Exim supports, including LDAP, Postgres, MySQL, etc, as long as the -result is a list of colon-separated strings; +result is a list of colon-separated strings. Several expansion variables are set before the DATA ACL is processed, and you can use them in this ACL. The following @@ -781,7 +679,10 @@ expansion variables are available: o $dmarc_status This is a one word status indicating what the DMARC library - thinks of the email. + thinks of the email. It is a combination of the results of + DMARC record lookup and the SPF/DKIM/DMARC processing results + (if a DMARC record was found). The actual policy declared + in the DMARC record is in a separate expansion variable. o $dmarc_status_text This is a slightly longer, human readable status. @@ -790,6 +691,11 @@ expansion variables are available: This is the domain which DMARC used to look up the DMARC policy record. + o $dmarc_domain_policy + This is the policy declared in the DMARC record. Valid values + are "none", "reject" and "quarantine". It is blank when there + is any error, including no DMARC record. + o $dmarc_ar_header This is the entire Authentication-Results header which you can add using an add_header modifier. @@ -825,6 +731,9 @@ b. Configure, somewhere before the DATA ACL, the control option to warn !domains = +screwed_up_dmarc_records control = dmarc_enable_forensic + warn condition = (lookup if destined to mailing list) + set acl_m_mailing_list = 1 + (DATA ACL) warn dmarc_status = accept : none : off !authenticated = * @@ -840,6 +749,10 @@ b. Configure, somewhere before the DATA ACL, the control option to set $acl_m_quarantine = 1 # Do something in a transport with this flag variable + deny condition = ${if eq{$dmarc_domain_policy}{reject}} + condition = ${if eq{$acl_m_mailing_list}{1}} + message = Messages from $dmarc_used_domain break mailing lists + deny dmarc_status = reject !authenticated = * message = Message from $domain_used_domain failed sender's DMARC policy, REJECT @@ -1015,6 +928,360 @@ Where SPAMMER_SET is a macro and it is defined as set acl_c_spam_host = ${lookup redis{GET...}} +Proxy Protocol Support +-------------------------------------------------------------- + +Exim now has Experimental "Proxy Protocol" support. It was built on +specifications from: +http://haproxy.1wt.eu/download/1.5/doc/proxy-protocol.txt +Above URL revised May 2014 to change version 2 spec: +http://git.1wt.eu/web?p=haproxy.git;a=commitdiff;h=afb768340c9d7e50d8e + +The purpose of this function is so that an application load balancer, +such as HAProxy, can sit in front of several Exim servers and Exim +will log the IP that is connecting to the proxy server instead of +the IP of the proxy server when it connects to Exim. It resets the +$sender_address_host and $sender_address_port to the IP:port of the +connection to the proxy. It also re-queries the DNS information for +this new IP address so that the original sender's hostname and IP +get logged in the Exim logfile. There is no logging if a host passes or +fails Proxy Protocol negotiation, but it can easily be determined and +recorded in an ACL (example is below). + +1. To compile Exim with Proxy Protocol support, put this in +Local/Makefile: + +EXPERIMENTAL_PROXY=yes + +2. Global configuration settings: + +proxy_required_hosts = HOSTLIST + +The proxy_required_hosts option will require any IP in that hostlist +to use Proxy Protocol. The specification of Proxy Protocol is very +strict, and if proxy negotiation fails, Exim will not allow any SMTP +command other than QUIT. (See end of this section for an example.) +The option is expanded when used, so it can be a hostlist as well as +string of IP addresses. Since it is expanded, specifying an alternate +separator is supported for ease of use with IPv6 addresses. + +To log the IP of the proxy in the incoming logline, add: + log_selector = +proxy + +A default incoming logline (wrapped for appearance) will look like this: + + 2013-11-04 09:25:06 1VdNti-0001OY-1V <= me@example.net + H=mail.example.net [1.2.3.4] P=esmtp S=433 + +With the log selector enabled, an email that was proxied through a +Proxy Protocol server at 192.168.1.2 will look like this: + + 2013-11-04 09:25:06 1VdNti-0001OY-1V <= me@example.net + H=mail.example.net [1.2.3.4] P=esmtp PRX=192.168.1.2 S=433 + +3. In the ACL's the following expansion variables are available. + +proxy_host_address The (internal) src IP of the proxy server + making the connection to the Exim server. +proxy_host_port The (internal) src port the proxy server is + using to connect to the Exim server. +proxy_target_address The dest (public) IP of the remote host to + the proxy server. +proxy_target_port The dest port the remote host is using to + connect to the proxy server. +proxy_session Boolean, yes/no, the connected host is required + to use Proxy Protocol. + +There is no expansion for a failed proxy session, however you can detect +it by checking if $proxy_session is true but $proxy_host is empty. As +an example, in my connect ACL, I have: + + warn condition = ${if and{ {bool{$proxy_session}} \ + {eq{$proxy_host_address}{}} } } + log_message = Failed required proxy protocol negotiation \ + from $sender_host_name [$sender_host_address] + + warn condition = ${if and{ {bool{$proxy_session}} \ + {!eq{$proxy_host_address}{}} } } + # But don't log health probes from the proxy itself + condition = ${if eq{$proxy_host_address}{$sender_host_address} \ + {false}{true}} + log_message = Successfully proxied from $sender_host_name \ + [$sender_host_address] through proxy protocol \ + host $proxy_host_address + + # Possibly more clear + warn logwrite = Remote Source Address: $sender_host_address:$sender_host_port + logwrite = Proxy Target Address: $proxy_target_address:$proxy_target_port + logwrite = Proxy Internal Address: $proxy_host_address:$proxy_host_port + logwrite = Internal Server Address: $received_ip_address:$received_port + + +4. Recommended ACL additions: + - Since the real connections are all coming from your proxy, and the + per host connection tracking is done before Proxy Protocol is + evaluated, smtp_accept_max_per_host must be set high enough to + handle all of the parallel volume you expect per inbound proxy. + - With the smtp_accept_max_per_host set so high, you lose the ability + to protect your server from massive numbers of inbound connections + from one IP. In order to prevent your server from being DOS'd, you + need to add a per connection ratelimit to your connect ACL. I + suggest something like this: + + # Set max number of connections per host + LIMIT = 5 + # Or do some kind of IP lookup in a flat file or database + # LIMIT = ${lookup{$sender_host_address}iplsearch{/etc/exim/proxy_limits}} + + defer message = Too many connections from this IP right now + ratelimit = LIMIT / 5s / per_conn / strict + + +5. Runtime issues to be aware of: + - The proxy has 3 seconds (hard-coded in the source code) to send the + required Proxy Protocol header after it connects. If it does not, + the response to any commands will be: + "503 Command refused, required Proxy negotiation failed" + - If the incoming connection is configured in Exim to be a Proxy + Protocol host, but the proxy is not sending the header, the banner + does not get sent until the timeout occurs. If the sending host + sent any input (before the banner), this causes a standard Exim + synchronization error (i.e. trying to pipeline before PIPELINING + was advertised). + - This is not advised, but is mentioned for completeness if you have + a specific internal configuration that you want this: If the Exim + server only has an internal IP address and no other machines in your + organization will connect to it to try to send email, you may + simply set the hostlist to "*", however, this will prevent local + mail programs from working because that would require mail from + localhost to use Proxy Protocol. Again, not advised! + +6. Example of a refused connection because the Proxy Protocol header was +not sent from a host configured to use Proxy Protocol. In the example, +the 3 second timeout occurred (when a Proxy Protocol banner should have +been sent), the banner was displayed to the user, but all commands are +rejected except for QUIT: + +# nc mail.example.net 25 +220-mail.example.net, ESMTP Exim 4.82+proxy, Mon, 04 Nov 2013 10:45:59 +220 -0800 RFC's enforced +EHLO localhost +503 Command refused, required Proxy negotiation failed +QUIT +221 mail.example.net closing connection + + +DSN Support +-------------------------------------------------------------- + +DSN Support tries to add RFC 3461 support to Exim. It adds support for +*) the additional parameters for MAIL FROM and RCPT TO +*) RFC complient MIME DSN messages for all of + success, failure and delay notifications +*) dsn_advertise_hosts main option to select which hosts are able + to use the extension +*) dsn_lasthop router switch to end DSN processing + +In case of failure reports this means that the last three parts, the message body +intro, size info and final text, of the defined template are ignored since there is no +logical place to put them in the MIME message. + +All the other changes are made without changing any defaults + +Building exim: +-------------- + +Define +EXPERIMENTAL_DSN=YES +in your Local/Makefile. + +Configuration: +-------------- +All DSNs are sent in MIME format if you built exim with EXPERIMENTAL_DSN=YES +No option needed to activate it, and no way to turn it off. + +Failure and delay DSNs are triggered as usual except a sender used NOTIFY=... +to prevent them. + +Support for Success DSNs is added and activated by NOTIFY=SUCCESS by clients. + +Add +dsn_advertise_hosts = * +or a more restrictive host_list to announce DSN in EHLO answers + +Those hosts can then use NOTIFY,ENVID,RET,ORCPT options. + +If a message is relayed to a DSN aware host without changing the envelope +recipient the options are passed along and no success DSN is generated. + +A redirect router will always trigger a success DSN if requested and the DSN +options are not passed any further. + +A success DSN always contains the recipient address as submitted by the +client as required by RFC. Rewritten addresses are never exposed. + +If you used DSN patch up to 1.3 before remove all "dsn_process" switches from +your routers since you don't need them anymore. There is no way to "gag" +success DSNs anymore. Announcing DSN means answering as requested. + +You can prevent Exim from passing DSN options along to other DSN aware hosts by defining +dsn_lasthop +in a router. Exim will then send the success DSN himself if requested as if +the next hop does not support DSN. +Adding it to a redirect router makes no difference. + + +Certificate name checking +-------------------------------------------------------------- +The X509 certificates used for TLS are supposed be verified +that they are owned by the expected host. The coding of TLS +support to date has not made these checks. + +If built with EXPERIMENTAL_CERTNAMES defined, code is +included to do so, and a new smtp transport option +"tls_verify_cert_hostname" supported which takes a list of +names for which the checks must be made. The host must +also be in "tls_verify_hosts". + +Both Subject and Subject-Alternate-Name certificate fields +are supported, as are wildcard certificates (limited to +a single wildcard being the initial component of a 3-or-more +component FQDN). + + +DANE +------------------------------------------------------------ +DNS-based Authentication of Named Entities, as applied +to SMTP over TLS, provides assurance to a client that +it is actually talking to the server it wants to rather +than some attacker operating a Man In The Middle (MITM) +operation. The latter can terminate the TLS connection +you make, and make another one to the server (so both +you and the server still think you have an encrypted +connection) and, if one of the "well known" set of +Certificate Authorities has been suborned - something +which *has* been seen already (2014), a verifiable +certificate (if you're using normal root CAs, eg. the +Mozilla set, as your trust anchors). + +What DANE does is replace the CAs with the DNS as the +trust anchor. The assurance is limited to a) the possibility +that the DNS has been suborned, b) mistakes made by the +admins of the target server. The attack surface presented +by (a) is thought to be smaller than that of the set +of root CAs. + +DANE scales better than having to maintain (and +side-channel communicate) copies of server certificates +for every possible target server. It also scales +(slightly) better than having to maintain on an SMTP +client a copy of the standard CAs bundle. It also +means not having to pay a CA for certificates. + +DANE requires a server operator to do three things: +1) run DNSSEC. This provides assurance to clients +that DNS lookups they do for the server have not +been tampered with. The domain MX record applying +to this server, its A record, its TLSA record and +any associated CNAME records must all be covered by +DNSSEC. +2) add TLSA DNS records. These say what the server +certificate for a TLS connection should be. +3) offer a server certificate, or certificate chain, +in TLS connections which is traceable to the one +defined by (one of?) the TSLA records + +There are no changes to Exim specific to server-side +operation of DANE. + +The TLSA record for the server may have "certificate +usage" of DANE_TA(2) or DANE_EE(3). The latter specifies +the End Entity directly, i.e. the certificate involved +is that of the server (and should be the sole one transmitted +during the TLS handshake); this is appropriate for a +single system, using a self-signed certificate. + DANE_TA usage is effectively declaring a specific CA +to be used; this might be a private CA or a public, +well-known one. A private CA at simplest is just +a self-signed certificate which is used to sign +cerver certificates, but running one securely does +require careful arrangement. If a private CA is used +then either all clients must be primed with it, or +(probably simpler) the server TLS handshake must transmit +the entire certificate chain from CA to server-certificate. +If a public CA is used then all clients must be primed with it +(losing one advantage of DANE) - but the attack surface is +reduced from all public CAs to that single CA. +DANE_TA is commonly used for several services and/or +servers, each having a TLSA query-domain CNAME record, +all of which point to a single TLSA record. + +The TLSA record should have a Selector field of SPKI(1) +and a Matching Type field of SHA2-512(2). + +At the time of writing, https://www.huque.com/bin/gen_tlsa +is useful for quickly generating TLSA records; and commands like + + openssl x509 -in -pubkey -noout /dev/null \ + | openssl sha512 \ + | awk '{print $2}' + +are workable for 4th-field hashes. + +For use with the DANE_TA model, server certificates +must have a correct name (SubjectName or SubjectAltName). + +The use of OCSP-stapling should be considered, allowing +for fast revocation of certificates (which would otherwise +be limited by the DNS TTL on the TLSA records). However, +this is likely to only be usable with DANE_TA. NOTE: the +default of requesting OCSP for all hosts is modified iff +DANE is in use, to: + + hosts_request_ocsp = ${if or { {= {0}{$tls_out_tlsa_usage}} \ + {= {4}{$tls_out_tlsa_usage}} } \ + {*}{}} + +The (new) variable $tls_out_tlsa_usage is a bitfield with +numbered bits set for TLSA record usage codes. +The zero above means DANE was not in use, +the four means that only DANE_TA usage TLSA records were +found. If the definition of hosts_require_ocsp or +hosts_request_ocsp includes the string "tls_out_tlsa_usage", +they are re-expanded in time to control the OCSP request. + +This modification of hosts_request_ocsp is only done if +it has the default value of "*". + + +For client-side DANE there are two new smtp transport options, +hosts_try_dane and hosts_require_dane. They do the obvious thing. +[ should they be domain-based rather than host-based? ] + +DANE will only be usable if the target host has DNSSEC-secured +MX, A and TLSA records. + +(TODO: specify when fallback happens vs. when the host is not used) + +If dane is in use the following transport options are ignored: + tls_verify_hosts + tls_try_verify_hosts + tls_verify_certificates + tls_crl + tls_verify_cert_hostnames + +Currently dnssec_request_domains must be active (need to think about that) +and dnssec_require_domains is ignored. + +If verification was successful using DANE then the "CV" item +in the delivery log line will show as "CV=dane". + +There is a new variable $tls_out_dane which will have "yes" if +verification succeeded using DANE and "no" otherwise (only useful +in combination with EXPERIMENTAL_TPDA), and a new variable +$tls_out_tlsa_usage (detailed above). + -------------------------------------------------------------- End of file