Wireshark Filters

Table of contents

1. Wireshark Filters For Beginners

2. Wireshark Filter Operators

3. Logical Wireshark Filter Operators

4. Interface Filters

5. Link layer traffic

6. Internet layer traffic

7. Transport layer traffic

8. Application layer traffic

9. Filters for Wi-Fi frames

Wireshark Filters For Beginners

Wireshark has a huge variety of different filters. And there is a huge documentation devoted to these filters. This documentation is not easy to understand. I have compiled the most interesting Wireshark Filters for me - for novice users this can become something like a reference for Wireshark filters. By the way, here in the comments you can share filters that you like - I will add them to this list.

Remember that Wireshark has display filters and capture filters. Here I consider the display filters that are entered in the main window of the program in the top field below the menu and icons of the main functions.

To fully understand the importance of filters and their meanings, it is necessary to understand how the network works.

Some filters are written here in a general form, and some are made as literal examples. Remember that in any case you can substitute your data, for example, change the port number to any one of your interest, and also do the same with the IP address, MAC address, time value, etc.

Wireshark Filter Operators

Filters can have different values, for example, it can be a string, a hexadecimal format or a number.

If an inaccurate entry is sought (better suited for non-numeric values) then contains is used. For example, to show TCP packets containing the string hackware you need the following filter:

tcp contains hackware

Operators are used to find exact values. Consider them:



==/eq Equals
!=/ne Not equal
</lt Less than
<=/le Less than or equal
>/gt More than
>=/ge Greater than or equal

As you can see, there are two versions, for example, if we want to indicate that the filter value is equal to something, then we can use == or eq.

Filters using logical operands can be used to build rather complex structures, but apparently, usage of the same filter with comparison operators is buggy. For example, as here in an attempt to filter not one port, but the range of ports:

tcp.port>=8000 && tcp.port<=8180

But actually the filter value (in this case, tcp.port) is overwritten by the last value, so as a result, instead of the expected behavior, we get the result of only the last part, in this case it is


Keep in mind this bug!

When using with == (equal), this bug is missing.

Logical Wireshark Filter Operators

Logical operators allow you to create detailed filters using several conditions at once. It is recommended to use brackets additionally, because otherwise you may not get the result you expect.

Operator Description
AND/&& Logical AND, data is output if they match to both parts of the filter. For example, the ip.src== and tcp filter will show only packets that originate from and which are associated with the TCP protocol. Only data matching both conditions will be displayed.
OR/|| Logical OR, it is enough that only one condition is true; if both are true, then this also fine. For example, the tcp.port==80 or tcp.port==8080 filter will show TCP packets that are related (are the source or destination) to port 80 or 8080.
NOT/! Boolean is NOT used when we want to exclude some packages. That is, all packets will be shown, except those that satisfy the condition following the NOT. For example, the filter !dns will show all packets except DNS.

Combination examples:

Show HTTP or DNS traffic:

http or dns

Show all traffic except ARP, ICMP and DNS:

!(arp or icmp or dns)

Interface Filters

Show packets only sent or received on the wlan0 interface:

frame.interface_name == "wlan0"

Link Layer Traffic

To show ARP traffic:


Show ARP protocol frames sent from device with MAC address 00:c0:ca:96:cf:cb:

arp.src.hw_mac == 00:c0:ca:96:cf:cb

Show ARP protocol frames sent from a device with an IP address

arp.src.proto_ipv4 ==

Show ARP protocol frames sent to a device with a MAC address 00:00:00:00:00:00 (this address is used when the protocol tries to find out the target MAC address. Another popular address that can confuse you is ff:ff:ff:ff:ff:ff, this is broadcast address, that is, messages with this address are for all devices on the local network):

arp.dst.hw_mac == 00:00:00:00:00:00

Show ARP protocol frames sent to the device having the IP address

arp.dst.proto_ipv4 ==

Show Ethernet traffic:


Show frames (in general, all frames, not just ARP, as it was in the previous examples) sent from a device that has the MAC address 00:c0:ca:96:cf:cb:

eth.src == 00:c0:ca:96:cf:cb

Show frames sent to device with MAC address 78:cd:8e:a6:73:be:

eth.dst == 78:cd:8e:a6:73:be

Internet Layer Traffic

IPv4 Protocol Filtering

Show IP traffic (this includes TCP, UDP, as well as application layer protocols DNS, HTTP - that is, almost everything except the data link layer protocols that do not use IP addresses for data transmission (in local Ethernet networks they use MAC addresses)):


More precisely, it means IPv4 traffic, which is usually called just IP (Internet Protocol).

Show traffic associated with a specific IP address (enter it instead of x.x.x.x). Packets will be shown in which this IP address is the source of the data OR the recipient:

ip.addr == x.x.x.x

Show traffic associated with these two IP addresses. According to the only possible situation, one of these addresses is the source, and the second is the destination address.

ip.addr == x.x.x.x && ip.addr == y.y.y.y

Show traffic originated from the host with the IP address

ip.src ==

Show traffic whose destination is the host with the IP address

ip.dst ==

Filter subnets and IP ranges in Wireshark

You can specify a subnet instead of a single IP address:

ip.addr ==

Filtering traffic sent from a specific IP range. If you need to filter out traffic whose source is the subnet, then use a filter of the form:

ip.src ==

Filtering traffic destined for sending to a specific IP range. If you need to filter traffic whose destination is a subnet, then use a filter of the form:

ip.dst ==

Please note, IP protocol operates with IP addresses, but does not operate with ports. Ports are part of the TCP and UDP protocols. IP protocol is only responsible for routing traffic between nodes.

IPv6 Protocol Filtering

Show IPv6 (Internet Protocol Sixth Version) traffic:


Filtering by IPv6 Address. To filter by IPv6 address, use the filter:

ipv6.addr == 2604:a880:800:c1::2ae:d001

Filtering Subnets and IPv6 Ranges in Wireshark

You can specify a subnet for filtering instead of a single IPv6 address:

ipv6.addr == 2604:a880:800:c1::2ae:d000/64

If you need to filter out traffic originating from a specific IPv6 address:

ipv6.src == 2604:a880:800:c1::2ae:d001

If you need to filter traffic sent to a specific IPv6 address:

ipv6.dst == 2604:a880:800:c1::2ae:d001

Filtering traffic sent from a specific IPv6 range. If you need to filter out traffic whose source is the subnet, then use a filter of the form:

ipv6.src == 2604:a880:800:c1::2ae:d000/64

Filtering traffic destined for sending to a specific IPv6 range. If you need to filter traffic whose destination is a subnet, then use a filter of the form:

ipv6.dst == 2604:a880:800:c1::2ae:d000/64

Filtering ICMPv6 (Internet Control Message Protocol of the sixth version) in Wireshark is done by the filter:


How to filter ARP packets for IPv6 in Wireshark? For IPv6, ARP is not required, since its role is played by the Neighbor Discovery Protocol (NDP) using ICPM6.

To see packets that act as ARP for IPv6, use a filter:

icmpv6.type == 133 or icmpv6.type == 134 or icmpv6.type == 135 or icmpv6.type == 136 or icmpv6.type == 137

Other filters with an IP address are similar for IPv6 and IPv4.

Transport Layer Traffic

To see only TCP traffic:


Show traffic whose source or destination port is a specific port, for example, 8080:


Show traffic originating from port 80:

tcp.srcport == 80

Show the traffic that is sent to the service listening on port 80:

tcp.dstport == 80

Show TCP packets with the SYN flag enabled:


Show TCP packets with the SYN flag enabled and the ACK flag disabled:

tcp.flags.syn==1 && tcp.flags.ack==0

Similarly for other flags:

  • SYN
  • ACK
  • RST
  • FIN
  • CWR
  • ECE
  • URG
  • PSH
  • NS

You can also use syntax like tcp.flags == 0x0XX, for example:

  • FIN is tcp.flags == 0x001
  • SYN is tcp.flags == 0x002
  • RST is tcp.flags == 0x004
  • ACK is tcp.flags == 0x010
  • ACK and FIN set simultaneously is tcp.flags == 0x011
  • ACK and SYN set simultaneously is tcp.flags == 0x012
  • ACK and RST set simultaneously is tcp.flags == 0x014

Data Offset:

tcp.hdr_len == 32
tcp.hdr_len == 52
tcp.hdr_len > 32

Packets with reserved bits set:

tcp.flags.res == 1

Window size:

tcp.window_size_value == 11
tcp.window_size_value == 4468
tcp.window_size_value > 65000
tcp.window_size_value < 100

Calculated window size:

tcp.window_size == 45056
tcp.window_size == 11

Window size scaling factor:

tcp.window_size_scalefactor == 4096

The tcp.window_size_value is the raw window size value as read directly from the TCP header, whereas tcp.window_size is the calculated window size, which is based on whether window scaling is applicable or not. If window scaling is not used or the scaling factor is 1 or if it's unknown whether window scaling is applicable or not because the TCP 3-way handshake was not captured, then the two values will be the same. You can tell by the tcp.window_size_scalefactor which of these conditions is applicable - if its value is -1, then it's unknown, if its value is -2, then window scaling is not used, and all other values represent an actual window scaling size factor.

To show packages containing any string, for example, the hackware string:

tcp contains hackware

Follow TCP stream number X:

tcp.stream eq X

Filter by stream number:

tcp.seq == x

Show transmitted packets. Helps track application performance and packet loss:


This filter displays problematic packets (lost segments, transmiting, etc.) TCP Keep-Alive packets pass this filter, but they are not an indication of problems.


Filters to assess the quality of a network connection.

The following specifications apply to TCP frames. Moreover, they are not based on frame headers – the considered characteristics (data omission, duplicates) are assigned by the Wireshark program based on the analysis.

The filter displays information about frames with the ACK flag, which are duplicates. A large number of such frames may be a sign of communication problems:

tcp.analysis.duplicate_ack_num == 1

Show frames for which the previous segment is not captured:


This is normal at the beginning of data capture – because the information is not intercepted from the very beginning of the session.

To display frames that are retransmission (resend):


Show frames that are not received in the correct order:


To see only UDP traffic:


Flags are not used for UDP. For this protocol, you can only specify the port.

Show traffic originating from port 53:

udp.srcport == 53

Show the traffic that is sent to the service listening on port 53:

udp.dstport == 53

A UDP packet containing a specific string, for example, a hackware string:

udp contains hackware

Destination Port OR Source Port:

udp.port == 53
udp.port > 40000
udp.port < 30

Package Length:

udp.length == 60
udp.length > 50000

Time between packets (to identify network problems):

udp.time_delta > 1.5

Stream Number (Request-Response):

udp.stream == 5

Traceroute packages:


To see only ICMP traffic:


To see only ICMP v6 traffic (sixth version)


Show all answers to ping:


Show all ping requests:


Show all unreachable/prohibited host and port errors


Show all attempts to redirect routing using ICMP:


Other ICMP types and CODE options are here: https://www.frozentux.net/iptables-tutorial/iptables-tutorial.html#TCPOPTIONS

An example of using the CODE value, the following filter will show messages about the port unreachable:

icmp.type == 3 && icmp.code == 3

Application layer traffic

For the application protocols of HTTP, DNS, SSH, FTP, SMTP, RDP, SNMP, RTSP, GQUIC, CDP, LLMNR, SSDP there are filters that are called like the protocols themselves, but are written in small letters.

For example, to see HTTP traffic:


To see the traffic of the new HTTP/2 protocol:


Remember that when deciding which protocol the transmitted data belongs to, the program considers the used port number. If a non-standard port is used, the program will not be able to find the necessary data. For example, if you connect to SSH on port 1234, the ssh filter will not find SSH traffic.

A filter that shows only the data sent by the POST method:

http.request.method == "POST"

A filter that shows only the data transmitted by the GET method:

http.request.method == "GET"

Search for requests to a specific site (host):

http.host == "<URL>"

Search requests to a specific site by part of the name:

http.host contains "here.particle.name"

Filter for outputting HTTP requests in which cookies were transmitted:


Requests in which the server has set cookies in the user's browser.


To search for any transferred images:

http.content_type contains "image"

To search for certain types of images:

http.content_type contains "gif"
http.content_type contains "jpeg"
http.content_type contains "png"

To search for files of a specific type:

http.content_type contains "text"
http.content_type contains "xml"
http.content_type contains "html"
http.content_type contains "json"
http.content_type contains "javascript"
http.content_type contains "x-www-form-urlencode"
http.content_type contains "compressed"
http.content_type contains "application"

Search for requests for files of a certain type. For example, to search for transferred ZIP archives:

http.request.uri contains "zip"

Instead of http.request.uri for greater accuracy, you can use the http.request.uri.path or http.request.uri.query filters, for example, to search for requests to download JPG files (links to pictures):

http.request.uri.path contains "jpg"

You can also filter requests that contain a specific HTTP REFERRER header value. For example, to search for queries in which the referrer is ru-board.com:

http.referer contains "ru-board.com"

Search requests with any authorization. Similarly, with the help of contains you can search for certain types of authorization:


Search for files in the HTTP stream:


To see which HTTP data is delayed, use the following construct:


It will show traffic received after 1 second.

To investigate problems, you can analyze the status of HTTP response codes. For example, the following filter will show traffic for which a 404 Not Found error was received (page not found):


The following filter is very interesting. Firstly, it shows which complex structures can be built from separate filters. Secondly, it allows you to explore HTTP requests and overall web activity, eliminating unnecessary data. With this filter you can view high level web activity. The rules inside the brackets exclude images, JavaScript files and style sheets — everything embeded the page requests within itself. If the examined pages contain other embedded objects, exclude them in a similar way:

http.request && !(http.request.uri contains ".ico" or http.request.uri contains ".css" or http.request.uri contains ".js" or http.request.uri contains ".gif" or http.request.uri contains ".jpg")

To see all DNS queries and answers:


To see which DNS queries took a lot of time:


The answers that came more than a second after sending the request will be shown.

This filter shows which dns requests cannot be correctly resolved:

dns.flags.rcode != 0

Show only DNS queries:

dns.flags.response == 0

Show only DNS responses:

dns.flags.response == 1

Show requests and responses in which the IP for google.com is searched:

dns.qry.name == "google.com"

Show DNS requests and responses for A record:

dns.qry.type == 1

Show DNS requests and responses for AAAA records:

dns.qry.type == 28

Show answers in which is sent for the A record as IP:

dns.a ==

Show replies in which 2a01:4f8:172:1d86::1 sent to the AAAA as IP

dns.aaaa == 2a01:4f8:172:1d86::1

Find certain value (apollo.archlinux.or) of CNAME:

dns.cname == "apollo.archlinux.org"

Show answers longer than 30:

dns.resp.len > 30

Show requests with a length of more than 25:

dns.qry.name.len >25

Show DNS server responses for which recursion is available:

dns.flags.recavail == 1

Show DNS server responses for which recursion is not available:

dns.flags.recavail == 0

Is recursion desirable (if the requested DNS server does not have host name information, should it query other DNS servers for this information):

dns.flags.recdesired == 1

If 1 is set, then recursion is needed, if 0 is set, then it is not desirable.

Whether to accept unauthenticated data (0 means do not accept, 1 means accept):

dns.flags.checkdisable == 0

To see how IP addresses are assigned by DHCP:


Or so:


To show DHCP requests:

bootp.option.dhcp == 3

To show DHCP Discover:

bootp.option.dhcp == 1

SMB filter. This filter in the Info column shows the entire Share of connections, open directories and open files in the trace.

smb2.cmd==3 or smb2.cmd==5

NetBIOS filters

Wireshark Filter for Name Service (NetBIOS-NS):


Wireshark Filter for Datagram Distribution Service (NetBIOS-DGM):


Wireshark Filter for Session Service (NetBIOS-SSN):



To show only VNC session traffic:


VNC challenge:


VNC response:


The result of an authentication attempt on the VNC server:


The challenge, response, and result of an authentication attempt on a VNC server:

vnc.auth_challenge or vnc.auth_response or vnc.auth_result

Computer Name with VNC Server:


Mouse button position:


Security Type Selected:


Encoding Type:


VNC client protocol version:


VNC server protocol version:


Framebuffer Width:




Key down:


Data of transmitted video in H.264 format:


Cut text:


Other VNC filters: https://www.wireshark.org/docs/dfref/v/vnc.html

Remote Desktop Protocol (RDP)

To show only RDP session traffic:


Other Remote Desktop Protocol (RDP) filters: https://www.wireshark.org/docs/dfref/r/rdp.html

Filters for Wi-Fi frames

Show elements of four-ways handshakes (i.e. EAPOL protocol frames):


Show Beacon frames:

wlan.fc.type_subtype == 0x08

Show Probe Response frames:

wlan.fc.type_subtype == 0x05

Show all at once: EAPOL, beacons, Probe Response:

wlan.fc.type_subtype == 0x08 || wlan.fc.type_subtype == 0x05 || eapol

Show wireless frames for a specific device with the BSSID MAC address:


Show EAPOL, Beacons, Probe Response for a specific device with a MAC address of 28:28:5D:6C:16:24:

(wlan.fc.type_subtype == 0x08 || wlan.fc.type_subtype == 0x05 || eapol) && wlan.addr==28:28:5D:6C:16:24

Show all PMKID:

eapol && wlan.rsn.ie.pmkid

Show PMKID, Beacons, Probe Response:

(wlan.fc.type_subtype == 0x08 || wlan.fc.type_subtype == 0x05 || (eapol && wlan.rsn.ie.pmkid))

Show PMKID, Beacons, Probe Response for access point with MAC address 40:3D:EC:C2:72:B8:

(wlan.fc.type_subtype == 0x08 || wlan.fc.type_subtype == 0x05 || (eapol && wlan.rsn.ie.pmkid)) && wlan.addr==40:3D:EC:C2:72:B8

Show only the first message of the handshake:

wlan_rsna_eapol.keydes.msgnr == 1

Show only the second message of the handshake (can be used for the handshake Message of any number):

wlan_rsna_eapol.keydes.msgnr == 2

Show frames for access points with a speed (Data Rate) of 1 Mb/s:

wlan_radio.data_rate == 1

Show frames for access points at speeds above 10 Mb/s:

wlan_radio.data_rate > 10

Show access points on a specific frequency:

radiotap.channel.freq == 2412

Show access points with a certain signal level:

wlan_radio.signal_dbm > -50

Filters associated with the presence of an antenna on a device:

radiotap.present.antenna == 1


radiotap.antenna == 1

If you know other interesting Wireshark filters, please share them in the comment section.

Last Updated on

Recommended for you:

Leave a Reply

Your email address will not be published.