Internet Engineering Task Force syslog Internet Draft: Informational Chris Lonvick draft-ietf-syslog-syslog-03.txt Cisco Systems January 3, 2001 Expires: July, 2001 syslog Protocol draft-ietf-syslog-syslog-03.txt STATUS OF THIS MEMO This document is an Internet-Draft and is in full conformance with all provisions of Section 10 of RFC 2026. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress". The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html. This work is a product of the IETF syslog Working Group. More information about this effort may be found at http://www.ietf.org/html.charters/syslog-charter.html Comments about this draft should be directed to the syslog working group at the mailing list of syslog-sec@employees.org. When written in uppercase, The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119. Copyright Notice Copyright (C) The Internet Society (2000). All Rights Reserved. Abstract This draft describes the observed behavior of the syslog protocol. This protocol has been used for the transmission of event notification messages across networks for many years. Expires July 2001 [Page 1] Draft syslog January 2001 1 Introduction Since the beginning, life has relied upon the transmission of messages. For the self-aware organic unit, these messages can relay many different things. The messages may signal danger, the presence of food or the other necessities of life, and many other things. In many cases, these messages are informative to other units and require no acknowledgement. As people created processes and machines, this same principle was applied to societal communications. As an example, severe weather warnings may be delivered through any number of channels - a siren blowing, warnings delivered over television and radio stations, and even through the use of flags on ships. The expectation is that people hearing or seeing these warnings would realize their significance and take appropriate action. In most cases, no responding acknowledgement of receipt of the warning is required or even desired. Along these same lines, operating systems, processes and applications were written to send messages of their own status, or messages to indicate that certain events had occurred. These event messages generally had local significance to the machine operators. As the operating systems, processes and applications grew ever more complex, systems were devised to categorize and log these diverse messages and allow the operations staff to more quickly differentiate the notifications of problems from simple status messages. The syslog process was one such system that has been widely accepted in many operating systems. Flexibility was designed into this process so the operations staff have the ability to configure the destination of messages sent from the processes running on the device. In one dimension, the events that were received by the syslog process could be logged to different files and also displayed on the console of the device. In another dimension, the syslog process could be configured to forward the messages across a network to the syslog process on another machine. The syslog process had to be built network-aware for some modicum of scalability since it was known that the operators of multiple systems would not have the time to access each system to review the messages logged there. The syslog process running on the remote devices could therefore be configured to either add the message to a file, or to subsequently forward it to another machine. In its most simplistic terms, the syslog protocol is an event notification protocol that allows a machine to send event notification messages across IP networks to event message collectors -also known as syslog servers. Since each process, application and operating system was written somewhat independently, there is little uniformity to the content of syslog messages. For this reason, no assumption is made upon the contents of the messages. The protocol is simply designed to transport these event messages. In all cases, there is one device that originates the message. The syslog process on that machine may send the message to a collector. No acknowledgement of the receipt is made. Expires July 2001 [Page 2] Draft syslog January 2001 1.1 Events and Generated Messages The writers of the operating systems, processes and applications have had total control over the circumstances that would generate any message. In some cases, messages are generated to give status. These can be either of a certain period of time, or at some other interval such as the invocation or exit of a program. In other cases, the messages may be generated due to a set of conditions being met. In that case, either a status message or a message containing an alarm of some type may be generated. The contents of a message have also been at the discretion of its creator. It has been considered to be good form to write the messages so that they are informative to the person who may be reading them. It has also been considered good practice to include a timestamp and some indication of the sending device and the process that originated it in the messages. However, none of those are required. It should be assumed that any process on any device might generate an event message. This may include processes on machines that do not have any local storage - e.g. printers, routers, hubs, switches, and diskless workstations. In that case, it may be imperative that event messages are transported to a collector so that they may be recorded and hopefully viewed by an operator. 1.2 Operations of the Message Receivers It is beyond the scope of this Internet Draft to specify how event messages should be processed. It was considered that the writers of the operating systems, processes and applications would quantify their messages into one of several broad categories. This was so that the operations staff could be presented with the more important and time sensitive messages quickly, while also having the ability to place status or informative messages in a file for later perusal. 2 Transport Layer Protocol syslog uses the user datagram protocol (UDP) [1] as its underlying transport layer mechanism. The UDP port that has been assigned to syslog is 514. It is RECOMMENDED that the source port also be 514 to indicate that the message is from the syslog process of the sender, but there have been cases seen where valid syslog messages have come from a sender with a source port other than 514. If the sender uses a source port other than 514 then it is RECOMMENDED and has been considered good form that subsequent messages are from a single consistent port. Expires July 2001 [Page 3] Draft syslog January 2001 3 Definitions and Architecture Any device may generate and send a syslog message. For the purpose of this document, we'll make the following definitions: - A machine that can generate a message will be called a "device". - A machine that can receive the message and relay it to another machine will be called a "relay". - A machine that recieves the message and does not relay it to any other machines will be called a "collector". This has been commonly known as a "syslog server". - Any device or relay will be known as the "sender" when it sends a message. - Any relay or collector will be known as the "receiver" when it receives the message. The archiecture of the devices may be summarized as follows: - Devices send messages to relays or collectors with no knowledge of whether it is a collector or relay. - Devices and relays may be configured to send the same message to multiple receivers. Expires July 2001 [Page 4] Draft syslog January 2001 The following architectures shown in Diagram 1 are valid while the first one has been known to be the most prevalent. +------+ +---------+ |Device|---->----|Collector| +------+ +---------+ +------+ +-----+ +---------+ |Device|---->----|Relay|---->----|Collector| +------+ +-----+ +---------+ +------+ +-----+ +-----+ +---------+ |Device|-->--|Relay|-->--..-->--|Relay|-->--|Collector| +------+ +-----+ +-----+ +---------+ +------+ +-----+ +---------+ |Device|---->----|Relay|---->----|Collector| | |-\ +-----+ +---------+ +------+ \ \ +-----+ +---------+ \-->--|Relay|---->----|Collector| +-----+ +---------+ +------+ +---------+ |Device|---->----|Collector| | |-\ +---------+ +------+ \ \ +-----+ +---------+ \-->--|Relay|---->----|Collector| +-----+ +---------+ +------+ +-----+ +---------+ |Device|---->----|Relay|---->-------|Collector| | |-\ +-----+ /--| | +------+ \ / +---------+ \ +-----+ / \-->--|Relay|-->--/ +-----+ Diagram 1. Some Possible syslog Architectures Pemutations of the above examples are also acceptable. As a very general rule, there are usually many devices sending messages to relatively fewer collectors. This fan-in process allows an administrator to aggregate messages into relatively few repositories. Expires July 2001 [Page 5] Draft syslog January 2001 4 Packet Format and Contents The syslog packet has two parts. The first part is the PRI and the second part is the MSG. The PRI has three, four, five, or six characters. The MSG will fill the remainder of the syslog packet. There is no ending delimiter but the total length of the packet MUST be 1024 bytes or less. There is no minimum length of the MSG although sending a syslog packet with no contents is worthless and SHOULD NOT be done. The MSG part of the packet has additional fields that are described in Section 4.2 below. 4.1 PRI Part of a syslog Packet The PRI part starts with a leading "<" ('less-than' character), followed by a number, which is followed by a ">" ('greater-than' character). This is OPTIONALLY followed by a single space character. The code set used in this part MUST be seven-bit ASCII in an eight-bit field as described in RFC 2234 [2]. These are the ASCII codes as defined in "USA Standard Code for Information Interchange" [3]. In this, the "<" character is defined as the Augmented Backus-Naur Form (ABNF) %d60, and the ">" character has ABNF value %d62. The number is known as the Priority code and represents both the Facility and Severity as described below. The Priority code consists of one, two, or three decimal integers (ABNF DIGITS) using values of %d48 (for "0") through %d57 (for "9"). The OPTIONAL space character at the end of this field is %d32. The Facilities and Severities of the messages are numerically coded with decimal values. The operating system and some of the daemons and processes have been assigned a Facility parameter. Processes and applications that have not been assigned a Facility, or that have not been configured to use one of the "local use" Facilities SHOULD use the "user" Facility which has the numerical Facility code of decimal 8. All Facilities are shown in the following table along with their numerical code values. Expires July 2001 [Page 6] Draft syslog January 2001 Numerical Facility Code 0 kernel messages 8 user-level messages 16 mail system 24 system daemons 32 security/authorization messages (note 1) 40 messages generated internally by syslogd 48 line printer subsystem 56 network news subsystem 64 UUCP subsystem 72 clock daemon (note 2) 80 security/authorization messages (note 1) 88 FTP daemon 96 NTP subsystem 104 log audit (note 1) 112 log alert (note 1) 120 clock daemon (note 2) 128 local use 0 (local0) 136 local use 1 (local1) 144 local use 2 (local2) 152 local use 3 (local3) 160 local use 4 (local4) 168 local use 5 (local5) 176 local use 6 (local6) 184 local use 7 (local7) Table 1. syslog Message Facilities Note 1 - Various operating systems have been found to utilize Facilities 32, 80, 112 and 120 for security/authorization, audit and alert messages which seem to be similar. Note 2 - Various operating systems have been found to utilize both Facilities 72 and 120 for clock (cron/at) messages. Expires July 2001 [Page 7] Draft syslog January 2001 Each message Priority also has a decimal Severity level indicator. These are described in the following table along with their numerical values. Numerical Severity Code 0 Emergency: system is unusable 1 Alert: action must be taken immediately 2 Critical: critical conditions 3 Error: error conditions 4 Warning: warning conditions 5 Notice: normal but significant condition 6 Informational: informational messages 7 Debug: debug-level messages Table 2. syslog Message Severities The Priority code is calculated by summing the numerical values of the codes of the Facility and Severity. For example, A kernel message with a Severity of Emergency would have a Priority code of decimal 0, while a "local use 4" message with a Severity of Notice would have a Priority code of decimal 165. 4.2 MSG Part of a syslog Packet The MSG part of the syslog packet MUST contain visible (printing) characters. The code set traditionally and most often used has also been seven-bit ASCII in an eight-bit field like that used in the PRI part. In this code set, the only allowable characters are the ABNF VCHAR values (%d21-126) and spaces (SP value %d20). However, no indication of the code set used within the MSG is required, nor is it expected. Other code sets MAY be used as long as the characters used in the MSG are exclusively visible characters and spaces similar to those described above. The selection of a code set used in the MSG SHOULD be made with thoughts of the intended receiver. A message containing characters in a code set that cannot be viewed by a recipient will yield no information of value to an operator or administrator looking at it. Any device receiving a syslog packet may change the format of the MSG before retransmitting it but only in ways specified in section 4.2.2. While a relay MUST adhere to the rules described in that section, it is beyond the scope of this memo to say how recieved packets are handled by a collector. It has generally been observed, however, that collectors will behave as a relay in the manner in which they will modify the MSG before displaying or recording them. Expires July 2001 [Page 8] Draft syslog January 2001 4.2.1 Original syslog Packets A device SHOULD compose a syslog packet with the PRI, a space character and then the MSG. There are no set requirements on the contents of the MSG as it is originally sent from a device. It is RECOMMENDED that the MSG have the following fields: TIMESTAMP, HOSTNAME, TAG and then the CONTENT. The contents of these fields are described here and their formats are detailed in Section 4.2.3. If the originally formed message has a TIMESTAMP, then it is the local time of the device. If the originally formed message has a HOSTNAME field, then it will contain the hostname as it knows itself. If it does not have a hostname, then it will contain its own IP address. If the originally formed message has a TAG value, then that will be the name of the program or process that generated the message. The CONTENT contains the details of the message. This has traditionally been a freeform message that gives some detailed information of the event. Even if nothing else is sent, the packet MUST contain something in the CONTENT field. 4.2.2 Relayed syslog Packets When a relay receives a packet, it will check for a valid PRI. If the first character is not a less-than sign, the relay MUST assume that the packet does not contain a valid PRI. If the 3rd, 4th, or 5th character is not a greater-than sign, the relay again MUST assume that the PRI is not valid. If the relay has been configured to forward packets with a Priority value of 14 (User Facility=8 and Informational Severity=6) then the relay MUST insert a PRI with a Priority value of 14 as well as a MSG as described below. The contents of the received packet will be treated as the CONTENT of the MSG and appended. This new packet will be sent to the next relay or collector. If the relay finds a valid PRI then it will check its internal configuration. Relays MUST be configured to forward syslog packets on the basis of their Priority value. If the relay finds that it is configured to forward the received packet, then it MUST check for a valid TIMESTAMP as the first field in the MSG. If it finds a valid TIMESTAMP, then it MUST relay the entire packet unchanged. However, if it does not find a valid TIMESTAMP, then it MUST add a TIMESTAMP. It SHOULD additionally add a HOSTNAME. These fields are described here and detailed in Section 4.2.3. The remainder of the packet MUST be treated as the CONTENT field of the MSG and appended. The TIMESTAMP will be the current local time of the relay. Expires July 2001 [Page 9] Draft syslog January 2001 The HOSTNAME will be the name of the device as it is known by the relay. If the name cannot be determined, the IP address of the device will be used. The Domain Name MUST NOT be included in the HOSTNAME field. If the relay adds a TIMESTAMP and HOSTNAME at the front of the MSG then it MUST check that the total length of the packet is still 1024 octets or less. If the packet has been expanded beyond 1024 octets, then the relay MUST truncate the packet to be 1024 octets. This may cause the loss of vital information from the end of original packets. It is for this reason that it is RECOMMENDED that the MSG part of syslog packets contain the fields documented in Section 4.2.1. 4.2.3 Formats of the Fields in the MSG The TIMESTAMP field is in the format of "MMM DD hh:mm:ss" (without the quote marks) where: MMM is the English language month of the year with the first character in uppercase and the other two characters in lowercase. The following are the only acceptable values: Jan, Feb, Mar, Apr, May, Jun, Jul, Aug, Sep, Oct, Nov, Dec DD is the day of the month. If the day of the month is less than 10, then it MUST be represented as a space and then the number. hh:mm:ss is the local time. This is represented in a 24-hour format in the hour field. The TIMESTAMP field is followed by a space character before the next field. The HOSTNAME field MUST contain the hostname of the device as specified in STD-13 [4]. This field is also followed with a space character before the next field. The TAG is a set of ABNF alpha-numeric string that MUST NOT exceed 32 characters. The TAG field will be terminated by any non-alpha- numeric character. While any non-alpha-numeric character will terminate this field, it is usually terminated by either the space character or the left square bracket character ("["). This is explained more in Section 5.3. The CONTEXT field may fill the remainder of the MESSAGE. Expires July 2001 [Page 10] Draft syslog January 2001 5 Conventions Although Section 4 of this document specifies all requirements for the syslog protocol format and contents, certain conventions have come about over time for the inclusion of additional information within the syslog message. It must be plainly stated that these datum are not mandated but may be included for completeness and to give the recipient some additional clues of their origin and nature. 5.1 Dates and Times It has been found that some network administrators like to archive their syslog messages over long periods of time. For the convenience of these people and for automated message parsers, a more explicit time stamp has been seen to have been added to some messages. Some devices will send an original syslog message with a 2 character or 4 character year field immediately after the space following the TIMESTAMP. Relays will only check the validity of the TIMESTAMP field and will not check the validity of the remaining fields. This is not consistent with the original intent of the order and format of the fields. If implementors wish to contain a more specific date and time stamp within the message, it should be within the CONTEXT field. Implementors may wish to utilize the ISO 8601 [5] date and time formats if they wish to include more explicit date and time information. 5.2 Domain Name and Address To readily identify the device that originated the message, it may be a good practice to include its fully qualified domain name (FQDN) and its IP address within the CONTEXT field. Traditionally, however, only the hostname has been included in the HOSTNAME field. 5.3 Originating Process Information It has also considered to be good practice to include some information about the process on the device that generated the message - if that concept exists. This is usually the process name and process id for robust operating systems. Some of the information is displayed in the TAG field. Quite often, additional information is included at the beginning of the CONTEXT field. The format of "TAG[pid]:" - without the quote marks - is common. The left square bracket is used to terminate the TAG field in this case. If the process id is immaterial, it may be left off. In that case, the TAG MAY be followed by a colon. This would be displayed as "TAG:" without the quotes. Expires July 2001 [Page 11] Draft syslog January 2001 5.4 Examples As examples, these are valid messages as they may be observed on the wire between two devices. In the following examples, each message starts with the less-than character but has been indented, with line breaks inserted for readability. Example 1 <37> Oct 11 16:00:15 mymachine su: 'su root' failed for lonvick on /dev/pts/8 This example shows an authentication error in an attempt to acquire additional privileges. It also shows the command attempted and the user attempting it. This was recorded as an original message from the device called mymachine. A relay receiving this would not make any changes before sending it along as it contains a properly formatted TIMESTAME field. The TAG value in this example is "su" and it has been terminated by the colon which is the first character of the CONTEXT field. Example 2 <14>Use the BFG! While this is a valid message, it has extraordinarily little useful information. Since it is a user-generated message, it is consistent that it is not associated with a process. However, it does not contain a timestamp or any indication of the source of the message. If this message is stored on paper or disk, subsequent review of the message will not yield anything of value. This example is obviously an original message from a device. A relay MUST make changes to the message as described in Section 4.2.2 before forwarding it. The resulting message is shown below. <14> Dec 7 05:16:32 quakesrvr Use the BFG! In this relayed message, a TIMESTAMP has been added along with a HOSTNAME. Subsequent relays will not make any further changes to this message. Example 3 <160> Aug 24 03:24:00 am CST 1987 mymachine myproc[10]: %% It's time to make the do-nuts. %% Ingredients: Mix=OK, Jelly=OK # Devices: Mixer=OK, Jelly_Injector=OK, Frier=OK # Transport: Conveyer1=OK, Conveyer2=OK # %% Expires July 2001 [Page 12] Draft syslog January 2001 This message does have a proper TIMESTAMP field in the message. A relay will not modify this message before sending it, however, the HOSTNAME and TAG fields are not consistent with the definitions in Section 4.2.1. The HOSTNAME field would be construed to be "am" and the TAG value would be "CST". It should be noted that the information contained in the CONTEXT of this example is not telemetry data, nor is it supervisory control or data acquisition information. Due to the security concerns listed in Section 6 of this document, information of that nature should probably not be conveyed across this protocol. Example 4 <0> 1990 Oct 22 08:22:59 TZ-6 scapegoat.dmz.example.org 10.1.2.3 sched[0]: That's All Folks! This example has sufficient date and time information as well as a fully qualified domain name (FQDN) [4] and IP address. It appropriately lists the process name as well as the process id that generated the message. The information given after those datum is limited. Due to the indicated severity of the event, the process may not have been able to gather anything more informative. It may have been fortunate to have generated and sent this message at all. This example is obviously an original message from a device. While it does have all of the information required, it MUST be modified by a relay since the first field is not a valid TIMESTAMP as described in Section 4.2.3. A relay will add a TIMESTAMP and HOSTNAME as follows. <0> Oct 22 08:23:00 scapegoat 1990 Oct 22 08:22:59 TZ-6 scapegoat.dmz.example.org 10.1.2.3 sched[0]: That's All Folks! 6 Security Considerations An odor may be considered to be a message that does not require any acknowledgement. People tend to avoid bad odors but are drawn to odors that they associate with good food. The acknowledgement of the receipt of the odor or scent is not required and indeed it may be the height of discretion to totally ignore some odors. On the other hand, it is usually considered good civility to acknowledge the prowess of the cook merely from the ambiance wafting from the kitchen. Similarly, various species have been found to utilize odors to attract mates. One species of moths use this scent to find each other. However, it has been found that bolas spiders can mimic the odor of the female moths of this species. This scent will then attract male moths which will follow it with the expectation of finding a mate. Instead, when they arrive at the source of the Expires July 2001 [Page 13] Draft syslog January 2001 scent, they will be eaten. [6] This is a case of a false message being sent out with inimical intent. In its local use, the syslog process places event notification messages into files on that system. This relies upon the integrity of the system for the protection of the messages. The subsequent configuration of the syslog process to use the syslog protocol to transport the messages to another collector was an extension of the delivery of event notification messages and exhibits the same trust of the network. As such there are some concerns about the applicability of this protocol in situations that require robust delivery. Along the lines of the analogy, computer event messages may be sent accidentally, erroneously and even maliciously. At the time of this writing, however, there have not been any reports of any machine consuming any other machine. 6.1 Packet Parameters As was described above, the message length MUST NOT exceed 1024 bytes. Attacks have seen where syslog messages are sent to a receiver that have message lengths greater than 1024 bytes. In some older versions of syslog, the receipt of syslog packets that had a message greater than 1024 bytes caused problems. syslog message receivers must not malfunction upon the receipt of packets where the message length is greater than 1024 bytes. If a message receiver does receive a message whose length is greater than 1024 bytes, it may log all of, or the source address and some of the contents of the message, or it may discard the message altogether. Devices MUST NOT retransmit messages whose received length exceeds 1024 bytes. Similarly, the receiver must rigidly enforce the correctness of the message body. syslog collectors must not malfunction if received messages do not have the less-than and greater-than characters around a valid Priority value. The receiver may locally log some or all of a received invalid message or they may discard it totally. Also, received messages must contain printable text in the message as was described in section 3. Devices must not malfunction if they receive a message containing characters other than the characters described above. 6.2 Message authenticity The syslog delivery mechanism does not strongly associate the message with the message sender. Any device can generate any syslog message and send it to any other machine through the syslog protocol. The receiver of that packet will not be able to ascertain that the message was indeed sent from the reported sender, or if the packet was sent from another device. Expires July 2001 [Page 14] Draft syslog January 2001 One possible consequence of this is that a misconfigured machine may send syslog messages to a collector representing itself as another machine. The administrative staff may become confused that the status of the supposed sender of the messages may not be accurately reflected in the received messages. The administrators may not be able to readily discern that there are two or more machines representing themselves as the same machine. Another consequence happens when the event messages are forwarded. Unless the identification of the device is contained within the body of the event message, the source of the message may be lost since it may only be self-identified by its IP address contained in the IP header. It should be noted that some cases of embedding the identity of a device may only have local significance and that may only be ephemeral. The inclusion of a fully qualified domain name in the message may give the administrators the best chance of identifying the source of each message if it can always be associated with an IP address. However, if the device had obtained an IP address from a DHCP pool, then that association would not always hold true. Malicious exploits of this vulnerability have also been noted. An attacker may transmit syslog messages (either from the machine that the messages purport from which to be sent or from any other machine) to a collector. The attacks that have been noted run along these lines: - An attacker may perform a Denial of Service attack by filling the disk of the collector with false messages, or otherwise overwhelming the collector by sending more messages than it can receive or process. - An attacker may hide the true nature of an attack amidst many other messages. As an example, an attacker may start generating messages indicating a problem on some machine. This may get the attention of the system administrators who will spend their time investigating the alleged problem. During this time, the attacker may be able to compromise a different machine, or a different process on the same machine. - An attacker may generate false syslog messages to give untrue indications of status or of events. As an example, an attacker may stop a critical process on a machine, which may generate a notification of exit. The attacker may subsequently generate a false notification that the process had been restarted from another machine already under the control of the attacker. The system administrators may accept that misinformation and not verify that the process had indeed been restarted. Expires July 2001 [Page 15] Draft syslog January 2001 6.3 Sequenced delivery As a general rule, the forensics of a network anomaly rely upon reconstructing the sequence of events. In a perfect world, the messages would be received on the syslog collector in their order of generation from the other devices and anyone looking at these records would have an accurate picture of the sequence of events. Unfortunately, the syslog process and protocol do not ensure ordered delivery. This section details some of the problems that may be encountered from this. 6.3.1 Single Source to a Destination The syslog records are usually presented (placed in a file, displayed on the console, etc.) in the order in which they are received. This is not always in accordance with the sequence in which they were generated. As they are transported across an IP network, some out of order receipt should be expected. This may lead to some confusion as messages may be received that would indicate that a process has stopped before it was started. This may be somewhat rectified if the originating process had timestamped or numbered each of the messages before transmission. To be as effective as possible, both the source of the message and the syslog collector should both timestamp the messages. In this, both the sending device as well as the collector should utilize the same authoritative time source. It should be remembered, however, that not all devices are capable of receiving time updates, and not all processes timestamp their messages. 6.3.2 Multiple Sources to a Destination In syslog, there is no concept of unified event numbering. Single devices are free to include a sequence number within the event message but that can hardly be coordinated between multiple devices. In such cases, multiple devices may report that they are each sending message number one. Again, this may be rectified somewhat if the sending devices utilize a timestamp from an authoritative source in their messages. As has been noted, however, even messages from a single device to a single collector may be received out of order. This situation is compounded when there are several devices configured to send their syslog messages to a single collector. Messages from one device may be delayed so that messages from another device are received by the collector even though the messages from the first were generated before the messages from the second. If there is no timestamp or sequence number, then the messages may be presented in the order in which they were received which may give an inaccurate view of the sequence of actual events. Expires July 2001 [Page 16] Draft syslog January 2001 6.3.3 Multiple Sources to Multiple Destinations The plethora of configuration options available to the network administrators may further skew the perception of the order of events. It is possible to configure a group of devices to send the status messages -or other informative messages- to one collector, while sending messages of relatively higher importance to another collector. Additionally, the messages may be sent to different files on the same collector. If the messages do not contain timestamps from the source, it may be difficult to order the messages if they are kept in different places. An administrator may not be able to determine if a record in one file occurred before or after a record in a different file. This may be somewhat alleviated by placing marking messages with a timestamp into all destination files. If these have coordinated timestamps, then there will be some indication of the time of receipt of the individual messages. 6.3.4 Replaying Also, without any sequence indication or timestamp, messages may be recorded and replayed at a later time. An attacker may record a set of messages that indicate normal activity of a machine. At a later time, that attacker may remove that machine from the network and replay the syslog messages to the collector. The administrators would find nothing unusual in the received messages and their receipt would falsely indicate normal activity of the machine. 6.4 Reliable Delivery As there is no mechanism within either the syslog process or the protocol to ensure delivery, and since the underlying transport is UDP, some messages may be lost. They may either be dropped through network congestion, or they may be maliciously intercepted and discarded. The consequences of the drop of one or more syslog messages cannot be determined. If the messages are simple status updates, then their non-receipt may either not be noticed, or it may cause an annoyance for the system operators. On the other hand, if the messages are more critical, then the administrators may not become aware of a developing and potentially serious problem. Messages may also be intercepted and discarded by an attacker as a way to hide unauthorized activities. 6.5 Message Integrity Besides being discarded, syslog messages may be damaged in transit, or an attacker may maliciously modify them. In the case of a packet containing a syslog message being damaged, there are various mechanisms built into the link layer as well as into the IP [7] and Expires July 2001 [Page 17] Draft syslog January 2001 UDP protocols which may detect the damage. A damaged IP packet may be discarded by an intermediary router [8]. Damage to a UDP packet may be detected by the receiving UDP module which may silently discard it. In any case, the original contents of the message will not be delivered to the collector. Additionally, if an attacker is positioned between the sender and collector of syslog messages, then he may be able to intercept and modify those messages in transit to hide unauthorized activities. 6.6 Message Observation While there are no strict guidelines pertaining to the event message format, most syslog messages are generated in human readable form with the assumption that capable administrators should be able to read them and understand their meaning. Neither the syslog protocol nor the syslog application has any mechanism to provide confidentiality of the messages in transit. In most cases passing the clear-text messages is a benefit to the operations staff if they are sniffing the packets off of the wire. The operations staff may be able to read the messages and associate them with other events seen from other packets crossing the wire to track down and correct problems. Unfortunately, an attacker may also be able to observe the human-readable contents of syslog messages. The attacker may then use the knowledge gained from those messages to compromise a machine or do other damage. 6.7 Message Prioritization and Differentiation While the processes that create the messages may signify the importance of the events through the use of the message Priority value, there is no distinct association between this value and the importance of delivery of the packet. As an example of this consider an application that generates two event messages. The first is a normal status message but the second could be an important message denoting a problem with the process. This second message would have an appropriately higher Severity value associated with the importance of that event. If the operators had configured that both of these messages be transported to a syslog collector then they would, in turn, be given to UDP for transmission. Under normal conditions, no distinction would be made between them and they would be transmitted in their order. Again, under normal circumstances, the receiver would accept syslog messages as they are received. If many devices are transmitting normal status messages, but one is transmitting an important event message, there is no inherent mechanism within the syslog protocol to prioritize the important message over the other messages. Expires July 2001 [Page 18] Draft syslog January 2001 On a case-by-case basis, device operators may find some way to associate the different levels with the quality of service identifiers. As an example, the operators may elect to define some linkage between syslog messages that have a specific Priority with a specific value to be used in the IPv4 Precedence field [7], the IPv6 Traffic Class octet [9], or the Differentiated Services field [10]. In the above example, the operators may have the ability to associate the status message with normal delivery while associating the message indicating a problem with a high reliability, low latency queue as it goes through the network. This would have the affect of prioritizing the essential messages before the normal status messages. Even with this hop-by-hop prioritization, this queuing mechanism could still lead to head of line blocking on the transmitting device as well as buffer starvation on the receiving device if there are many near-simultaneous messages being sent or received. In this same line, some implementations of the syslog application may have mechanisms for the prioritization of the more important messages within the transmission queue. This behavior is not unique to syslog but is endemic to all operations that transmit messages serially. There are security concerns for this behavior. Head of line blocking of the transmission of important event messages may relegate the conveyance of important messages behind less important messages. If the queue is cleared appropriately, this may only add seconds to the transmission of the important message. On the other hand, if the queue is not cleared, then important messages may not be transmitted. Also at the receiving side, if the syslog receiver is suffering from buffer starvation due to large numbers of messages being received near-simultaneously, important messages may be dropped indiscriminately along with other messages. While these are problems with the devices and their capacities, the protocol security concern is that there is no prioritization of the relatively more important messages over the less important messages. 6.8 Misconfiguration Since there is no control information distributed about any messages or configurations, it is wholly the responsibility of the network administrator to ensure that the messages are actually going to the intended recipient. Cases have been noted where devices were inadvertently configured to send syslog messages to the wrong receiver. In many cases, the inadvertent receiver may not be configured to receive syslog messages and it will probably discard them. In certain other cases, the receipt of syslog messages has been known to cause problems for the unintended recipient. [11] If messages are not going to the intended recipient, then they cannot be reviewed or processed. Expires July 2001 [Page 19] Draft syslog January 2001 6.9 Forwarding Loop As it is noted in Figure 1, machines may be configured to relay syslog messages to subsequent relays before reaching a collector. In one particular case, an administrator found that he had mistakenly configured two relays to forward messages with certain priority values to each other. When either of these machines either received or generated that type of message, it would forward it to the other relay. That relay would, in turn, forward it back. This cycle did cause degradation to the intervening network as well as to the processing availability on the two devices. Network administrators must take care to not cause such a death spiral. 7 Conclusion and Other Efforts The syslog protocol may be effectively used to transport event notification messages across a network. It is highly recommended that the network operators who choose to use this understand the characteristics of the protocol and its security implications. There have been attempts in the past to standardize the format of the syslog message. The most notable attempt culminated in a BOF at the Fortieth Internet Engineering Task Force meeting in 1997. This was the Universal Logging Protocol (ulp) BOF and the minutes of their meeting is on-line at the IETF Proceedings web site. [12] Many good thoughts came from that effort and interested implementers may want to find some of the notes or papers produced from that effort. Expires July 2001 [Page 20] Draft syslog January 2001 8 Acknowledgements The following people provided content feedback during the writing of this draft: Jon Knight Magosanyi Arpad Balazs Scheidler Jon Callas Eliot Lear Petter Reinholdtsen Darren Reed Alfonso De Gregorio Eric Allman Eric Allman is the original inventor and author of the syslog daemon and protocol. The author of this draft and the community at large would like to express their appreciation for this work and for the usefulness that it has provided over the years. A large amount of additional information about this de-facto standard operating system feature may usually be found in the syslog.conf file as well as in the man pages for syslog.conf, syslog, syslogd, and logger, of many Unix and Unix-like devices. 9 Bibliography [1] Postel, J., "User Datagram Protocol", STD 6, RFC 768, USC/Information Sciences Institute, August 1980. [2] Crocker, D., and P. Overell, "Augmented BNF for Syntax Specifications: ABNF", RFC 2234, November, 1997 [3] USA Standard Code for Information Interchange, USASI X3.4-1968. [4] Mockapetris, P.V., "Domain names - concepts and facilities", RFC 1034, STD 13, Nov 1987. [5] Data elements and interchange formats - Information exchange - Representation of dates and times, International Organization for Standardization, Reference number ISO 8601 : 1988 (E), 1988 [6] Stowe, M., et al, "Chemical Mimicry: Bolas Spiders Emit Components of Moth Prey Species Sex Pheromones", Science, 1987 [7] Postel, J., "Internet Protocol", STD 5, RFC 791, USC/Information Sciences Institute, September 1981. [8] Baker, F., "Requirements for IP Version 4 Routers", RFC 1812, June 1995. Expires July 2001 [Page 21] Draft syslog January 2001 [9] Deering, S. and R. Hinden, "Internet Protocol, Version 6 (IPv6) Specification", RFC 2460, December 1998. [10] Nichols, K., S. Blake, F. Baker, D. Black, "Definition of the Differentiated Services Field (DS Field) in the IPv4 and IPv6 Headers", RFC 2474, December 1998 [11] Cisco Systems Product Security Incident Response Team (PSIRT), "Field Notice: Cisco IOS(r) Syslog Crash", January 11, 1999 http://www.cisco.com/warp/public/707/advisory.html [12] Walker, D., IETF Secretariat, "Proceedings of the Fortieth Internet Engineering Task Force, Washington, DC, USA, December 8-12, 1997 http://www.ietf.org/proceedings/97dec/index.html Expires July 2001 [Page 22] Draft syslog January 2001 A Author's Address Chris Lonvick Cisco Systems 12515 Research Blvd. Austin, TX 78759 USA +1.512.378.1182 clonvick@cisco.com B Full Copyright Statement Copyright (C) The Internet Society (2000). All Rights Reserved. This document and translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or assist in its implementation may be prepared, copied, published and distributed, in whole or in part, without restriction of any kind, provided that the above copyright notice and this paragraph are included on all such copies and derivative works. However, this document itself may not be modified in any way, such as by removing the copyright notice or references to the Internet Society or other Internet organizations, except as needed for the purpose of developing Internet standards in which case the procedures for copyrights defined in the Internet Standards process must be followed, or as required to translate it into languages other than English. The limited permissions granted above are perpetual and will not be revoked by the Internet Society or its successors or assigns. This document and the information contained herein is provided on an "AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK FORCE DISCLAIMS ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Expires July 2001 [Page 23]