Learn

284 articlesCategory: All
Network

What are communication logs?

Internet communication does not necessarily disappear completely on the spot.

Records about communication may remain on websites, servers, DNS, ISPs, routers, firewalls, and similar systems.

These records of communication are generally called "communication logs."

Communication logs may include not only the content of a communication, but also surrounding information such as communication time, source IP address, destination, accessed URL, browser information, cookies, and DNS queries.

When thinking about anonymity, it is not enough to ask only whether the content is encrypted.

That is because even when the content cannot be read, metadata associated with the communication can become a clue for tracking.

Where communication logs remain

Communication logs are records about communication.

For example, when you access a website, the web server may record "when," "from which IP address," "to which URL," and "with what kind of browser information" the access occurred.

However, communication logs do not remain in the same way in every environment. Which information is recorded changes depending on server settings, application design, cloud services used, network configuration, and log retention policy.

The important point is that internet communication has multiple observation points. Different kinds of logs may remain in different places, such as the website, DNS, ISPs, routers, internal networks, and cloud infrastructure.

Where logs may remainInformation that may be recordedMain purpose
Web serverAccess time, IP address, URL, User-Agent, referrer, and similar informationAccess analysis, incident response, attack detection
ApplicationLogin history, operation history, error details, account ID, and similar informationAbuse prevention, service operation
DNS resolverQueried domain name, query time, requester information, and similar informationName resolution, network management
ISP or network operatorConnection time, assigned IP address, traffic volume, some destination information, and similar informationNetwork operation, connection management
Router and firewallDestination, traffic volume, blocked communication, internal device information, and similar informationNetwork management, security measures

Why logs remain

Logs do not exist only to monitor users. In many cases, logs are recorded to operate services and networks stably.

When an error occurs on a website, administrators check logs to investigate which URL, at what time, and what kind of error occurred.

Logs also become important evidence for investigation when suspicious events occur, such as unauthorized login, heavy access, vulnerability scans, malware infection, or suspected information leakage.

PurposeHow logs are usedExample
Incident responseInvestigate causes of errors or outagesCheck whether 500 errors are increasing on a specific URL
Security measuresDetect suspicious accessCheck whether many login attempts occurred in a short time
Abuse preventionInvestigate clues to terms violations or attacksCheck abnormal requests or unnatural access sources
Service improvementUnderstand usageCheck frequently viewed pages and high-access time slots

Logs are a basic mechanism for protecting services. At the same time, because logs may contain information about user behavior, they can also become a privacy risk depending on how they are managed.

Logs that remain on websites

When you access a website, access logs may remain on the web server.

Typical access logs include source IP address, access time, HTTP method, URL, status code, User-Agent, referrer, and similar information.

An example URL is as follows.

URL example : https[:]//example.com/article/network-log

Here, example.com is an example domain often used for explanation.

On the website, even when communication uses HTTPS, the site receives the accessed URL path and query, cookies, and sent request information. HTTPS is a mechanism that makes content harder to read on the communication path; it does not prevent the destination website itself from receiving request information.

However, information received by the web server and information actually stored as logs are not the same. s, request bodies, and similar data may or may not be recorded depending on server and application settings.

InformationContentCaution
Access timeWhen access occurredMay become a clue matched with other logs by time
Source IP addressWhich IP address connectedThrough a CDN or proxy, the direct source may be an intermediary server
URLWhich page or API was accessedQuery strings may include identifiers
User-AgentBrowser, OS, app type, and similar informationMay become material for inferring the usage environment
CookieIdentifying information the site stores in the browser and receives from itMay be used to identify return visits from the same browser or login state
ReferrerWhich page the user moved fromMay not be sent, or only part may be sent, due to Referrer-Policy and similar mechanisms
Status codeRequest resultProcessing result is visible from 200, 301, 403, 404, 500, and similar codes

When a website uses a CDN, load balancer, or reverse proxy, the source IP address visible to the web server may be the IP address of the intermediary server rather than the actual user. In this case, the original client IP address may be recorded in headers such as X-Forwarded-For or in CDN-side logs.

Logs that remain in applications and authentication systems

Server-side logs are not limited to web server access logs.

Records about user operations and system behavior may also remain in login functions, payment functions, admin screens, APIs, databases, error monitoring tools, and similar systems.

For example, a service with a login function may record information such as the following.

  • Successful login
  • Failed login
  • Password change
  • Two-factor authentication attempt
  • Session issuance
  • Account settings change
  • Access to an admin screen

Services that provide APIs may record which API endpoint received a request, from which account or IP address, and at what time.

These logs are used to detect unauthorized login, account takeover, permission abuse, abnormal API use, and similar problems.

Records that may remain with ISPs and network operators

When connecting to the internet, user communication usually passes through the network of an ISP or access provider.

On the provider side, records about connections may remain, such as when a contracted line was connected, which IP address was assigned, and how much traffic occurred.

On home lines and mobile lines, the IP address assigned to a user may not be fixed and may change over time. Therefore, information about "which contracted line was using this IP address at this time" relates to records on the provider side.

Also, on mobile lines and some other lines, a system may be used where many users share one global IP address. In this case, not only the IP address but also information such as time and port number may relate to identifying the connection.

However, the existence of provider records and anyone being able to freely view that information are separate matters. ISP and network-operator logs are usually not something ordinary users can freely inspect.

Records related to DNS

When accessing a website, the browser or OS needs to convert the domain name into an IP address. This mechanism is called DNS.

For example, when accessing the following URL, the device looks up the IP address corresponding to example.com.

URL example : https[:]//example.com

DNS query records may retain which terminal or network made a query for which domain name.

DNS logs do not record the body text of a web page itself. However, they may show information about "which domain someone tried to access."

One caution is that where DNS queries are recorded changes depending on the environment. If you use your ISP's DNS, the query may reach the provider's DNS resolver. On the other hand, if the browser or OS uses another DNS resolver or encrypted DNS, the query destination changes.

In other words, when thinking about DNS logs, you need to look at "which DNS resolver was queried."

Router and firewall logs

Communication logs do not remain only with websites and ISPs.

Records about communication may also remain on home routers, company or school network equipment, cloud firewalls, proxy servers, gateways, and similar systems.

In organizational networks, records may be kept showing which device connected to which external server and when. These are used to investigate malware infection, unauthorized access, information leakage, insider abuse, and similar issues.

On the server side as well, multiple logs may exist, such as OS authentication logs, firewall logs, SSH logs, application logs, and cloud audit logs.

For example, the following kinds of logs are important in server administration.

Log typeContent that may be recordedHow it is used
Authentication logSuccessful login, failed login, source IP address, and similar informationChecking unauthorized login and brute-force attacks
Firewall logAllowed and blocked communication, port number, destination, and similar informationDetecting suspicious communication and attacks
Proxy logAccess from internal devices to external sitesInternal communication management and investigation
Cloud audit logAdministrative operation, API execution, permission change, and similar informationTracking setting changes and permission abuse
Application logErrors, operation history, processing results, and similar informationIncident response and abuse investigation

In this way, communication logs are not only "records of accessing a website"; they exist across networks and systems as a whole.

Relationship between HTTPS and logs

HTTPS is an important mechanism for encrypting content in transit and confirming that the communication partner is the intended party.

However, using HTTPS does not mean that no one can know any communication information.

The main target protected by HTTPS is content that could otherwise be visible to third parties on the communication path. For example, form input, page body, many HTTP headers, URL paths and queries, and similar content normally become harder to read on the communication path.

On the other hand, the destination web server decrypts the communication and processes the request. Therefore, the website receives the accessed URL, cookies, User-Agent, login information, submitted form content, and similar information.

Also, even on the communication path, connection time, source IP address, destination IP address, traffic volume, and some information about the TLS connection may be observable.

In other words, HTTPS is very important, but it is not a mechanism that erases all log issues.

ObserverInformation that may be visibleInformation harder to see
Website sideURL, cookies, User-Agent, submitted content, login operations, and similar informationContent you do not want third parties on the communication path to see
Third party on the communication pathSource IP, destination IP, communication time, traffic volume, and similar informationPage body, form content, URL path and query, and similar information
DNS resolverQueried domain name, query time, and similar informationWeb page body and specific operation details
Router and firewallDestination, traffic volume, allow/block result, and similar informationBody text inside HTTPS and many HTTP headers

Logs are not necessarily bad

Communication logs are information that needs attention from the perspective of anonymity and privacy. However, logs themselves are not bad.

Without logs, it becomes difficult to investigate server outage causes, detect unauthorized access, respond to account takeover, investigate malware infection, and operate services stably.

What matters is why logs are stored, what scope is recorded, how long they are stored, and who can access them.

The existence of logs and anyone being able to view them freely are separate matters. Properly managed logs are important information for protecting services, but excessively stored logs or poorly managed logs become privacy risks.

Logs become clues for anonymity

When thinking about anonymity, it is insufficient to look only at whether the content is encrypted.

Even when content cannot be read, information such as communication time, source IP address, destination, DNS queries, cookies, User-Agent, and referrer may be recorded in other forms.

These pieces of information may not directly identify an individual by themselves. However, when multiple records are combined, they may become material for inferring whether records appear to come from the same user, the access route, the behavioral timeline, the source network, and similar information.

InformationWhat may be learnedAnonymity caution
Access timeWhen communication occurredMay be matched with other records by time
IP addressWhich network connectedMay become material for inferring line, region, organization, and similar context
DNS queryWhich domain someone tried to accessBecomes a clue to the destination even if the page body is not known
CookieWhether access came from the same browserMay be used to identify return visits or login state
User-AgentBrowser or OS typeMay be treated as a feature of the usage environment
ReferrerWhich page the user came fromPart of the movement route may be known
URL queryMay include search terms, identifiers, session information, and similar dataIf it remains in server-side logs, it becomes a clue to behavior content

For anonymity, "the content is not read" and "communication traces do not remain" are not the same.

Even when content is encrypted, surrounding information about the communication may remain. For that reason, thinking about anonymity requires organizing not only encryption, but also where and what kinds of logs may remain.

Summary

Communication logs are records about communication.

Various kinds of information about communication may remain on websites, applications, DNS, ISPs, routers, firewalls, cloud infrastructure, and similar systems.

Representative information includes access time, source IP address, URL, User-Agent, cookie, referrer, DNS query, destination, traffic volume, authentication history, and operation history.

Logs are used for incident response, security measures, abuse investigation, service improvement, and similar purposes. Therefore, logs themselves are an important mechanism for operating internet services safely.

At the same time, from the perspective of anonymity, logs may become clues for tracking. That is because even when content is encrypted, source, destination, time, DNS queries, cookies, User-Agent, and similar information may be recorded in other forms.

Understanding the basics of communication logs makes it easier to organize what may be recorded on the internet. And when thinking about anonymity, it becomes possible to consider not only whether content is visible, but also where surrounding information about the communication may remain.

Related articles