Firewall / Spamassassin: Automatic list update

[imc67]

@necrevistonnezr said in Firewall / Spamassassin: Automatic list update:

I guess something like this should work

Did you managed it to get it worked like that?

[necrevistonnezr]

@imc67 No, as I don’t know how to avoid duplicates in the database - I don’t want to fill up a database with a hard entry limit with nonsense

[imc67]

@necrevistonnezr I gave it a try with some help by ChatGPT and it works flawless!! Except the API can't handle large list where the GUI is able to handle without an issue.

The script automatically downloads all the geo lists in an array, creates a copy/paste file for the GUI and then prepares the file in JSON style and connect/upload via API.

When I choose only a few countries is works perfect, however when choosing all the desired ones:

@girish @nebulon I get a line 83: /usr/bin/curl: Argument list too long

I can copy/paste the full list in the GUI, it takes some time but it uploads and settles all IP ranges (about 87k)

Does anyone know how to do this via the API?

[robi]

@imc67 Since it works from the browser, you should be able to inspect the API calls in the browser console.

[girish]

@imc67 said in Firewall / Spamassassin: Automatic list update:

@girish @nebulon I get a line 83: /usr/bin/curl: Argument list too long

this is related to curl . I don't know the answer but you can look for posts similar to https://stackoverflow.com/questions/54090784/curl-argument-list-too-long where you have to pass the args as a file instead of on the command line itself.

[imc67]

@girish said in Firewall / Spamassassin: Automatic list update:

https://stackoverflow.com/questions/54090784/curl-argument-list-too-long

I tried that in the beginning but got this error:

{
  "status": "Bad Request",
  "message": "blocklist must be a string"
}

[imc67]

@robi I did a check and somehow the browser is able to do a POST with the huge string ...

[imc67]

Again ChatGPT did it!!! It's really unbelievable how "patient" it is and after keep trying all it's variants (after feedback the errors) and asking if there is another way except curl it came with wget and after the second try it worked!

[imc67]

use at your own risk and be aware that with this long list it takes 1,5 minute to process the API call (same as in the GUI), replace the API input and copy/paste it in a .sh file (don't forget to make it executable). It's works but it's not perfect.

#!/bin/bash

# Huidige datum en tijd
current_datetime=$(date +"%Y%m%d_%H%M%S")

# Array met de URL's van de IP-lijsten en hun beschrijvingen
declare -a urls=(
	"https://iplists.firehol.org/files/spamhaus_drop.netset,Spamhaus - Drop"
        "https://iplists.firehol.org/files/spamhaus_edrop.netset,Spamhaus - eDrop"
        "https://www.ipdeny.com/ipblocks/data/aggregated/af-aggregated.zone,AF - Afganistan"
        "https://www.ipdeny.com/ipblocks/data/aggregated/bd-aggregated.zone,BD - Bangladesh"
        "https://www.ipdeny.com/ipblocks/data/aggregated/by-aggregated.zone,BY - Belarus"
        "https://www.ipdeny.com/ipblocks/data/aggregated/br-aggregated.zone,BR - Brazil"
        "https://www.ipdeny.com/ipblocks/data/aggregated/cn-aggregated.zone,CN - China"
        "https://www.ipdeny.com/ipblocks/data/aggregated/ir-aggregated.zone,IR - Iran"
        "https://www.ipdeny.com/ipblocks/data/aggregated/in-aggregated.zone,IN - India"
        "https://www.ipdeny.com/ipblocks/data/aggregated/kp-aggregated.zone,KP - North Korea"
        "https://www.ipdeny.com/ipblocks/data/aggregated/np-aggregated.zone,NP - Nepal"
        "https://www.ipdeny.com/ipblocks/data/aggregated/pk-aggregated.zone,PK - Pakistan"
        "https://www.ipdeny.com/ipblocks/data/aggregated/ro-aggregated.zone,RO - Romania"
        "https://www.ipdeny.com/ipblocks/data/aggregated/ru-aggregated.zone,RU - Russia"
        "https://www.ipdeny.com/ipblocks/data/aggregated/sg-aggregated.zone,SG - Singapore"
        "https://www.ipdeny.com/ipblocks/data/aggregated/sy-aggregated.zone,SY - Syria"
        "https://www.ipdeny.com/ipblocks/data/aggregated/tr-aggregated.zone,TR - Turkey"
        "https://www.ipdeny.com/ipblocks/data/aggregated/ua-aggregated.zone,UA - Ukraine"
        "https://www.ipdeny.com/ipblocks/data/aggregated/vn-aggregated.zone,VN - Vietnam"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/af-aggregated.zone,AF - Afganistan"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/bd-aggregated.zone,BD - Bangladesh"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/by-aggregated.zone,BY - Belarus"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/br-aggregated.zone,BR - Brazil"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/cn-aggregated.zone,CN - China"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/ir-aggregated.zone,IR - Iran"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/in-aggregated.zone,IN - India"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/np-aggregated.zone,NP - Nepal"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/pk-aggregated.zone,PK - Pakistan"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/ro-aggregated.zone,RO - Romania"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/ru-aggregated.zone,RU - Russia"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/sg-aggregated.zone,SG - Singapore"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/sy-aggregated.zone,SY - Syria"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/tr-aggregated.zone,TR - Turkey"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/ua-aggregated.zone,UA - Ukraine"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/vn-aggregated.zone,VN - Vietnam"
)

# Bestandsnaam met de huidige datum en tijd
output_file="samengevoegde_lijst_${current_datetime}.txt"

# Downloaden en samenvoegen van de IP-lijsten
for url_info in "${urls[@]}"
do
    # Opsplitsen van de URL-informatie
    IFS=',' read -r url description <<< "$url_info"

    # Opmerking toevoegen met de URL en beschrijving
    echo "# URL: $url" >> "$output_file"
    echo "# Description: $description" >> "$output_file"

    echo "IP-lijst downloaden van $url"
    # Downloaden van de IP-lijst en toevoegen aan het bestand
    curl -sS "$url" >> "$output_file"
done

echo "Samenvoegen voltooid! De samengevoegde lijst is opgeslagen in $output_file"

# Formatteren van het bestand voor de Cloudron Blocklist API
formatted_file="formatted_$output_file"

# Voeg "\n" toe aan het einde van elke regel
awk '{printf "%s\\n",$0}' "$output_file" > "$formatted_file"

# Cloudron Blocklist API endpoint
cloudron_api_endpoint="https://your-cloudron-domain.com/api/v1/network/blocklist"

# API Key voor authenticatie (vervang 'your-api-key' door jouw API-sleutel)
api_key="your-api-key"

# Uploaden naar Cloudron Blocklist API met wget
echo "Uploaden naar Cloudron Blocklist API met wget..."

# De gegevens in het vereiste formaat voor de API
data="{\"blocklist\":\"$(cat "$formatted_file" | tr '\n' '\\n')\"}"

# Verzend het bestand met een POST-verzoek via wget
echo "$data" > temp_data.txt
wget --method=POST --header="Content-Type: application/json" --header="Authorization: Bearer $api_key" --body-file=temp_data.txt "$cloudron_api_endpoint" --quiet --output-document=output.txt

# Toon de uitvoer van wget
cat output.txt

# Verwijder tijdelijke bestanden
rm temp_data.txt

[necrevistonnezr]

@imc67 thanks for this! Have you seen duplicate entries if you apply the method more than once?

[imc67]

Every time you execute the script the existing list is replaced by the newly generated version. There might be duplicates at generating the list because IP’s from the first two URL’s can exist in the later ones but that’s not a problem for me.

[necrevistonnezr]

Thanks again!
I just tried this - it generated merged_list_20240110_091244.txt and formatted_merged_list_20240110_091244.txt but somehow, the temp_data.txt is not generated, and therefore the upload is not successful?

BTW, here's your script with english comments and filenames:

#!/bin/bash

# Current date and time
current_datetime=$(date +"%Y%m%d_%H%M%S")

# Array containing the URLs of the IP lists and their descriptions
declare -a urls=(
	"https://iplists.firehol.org/files/spamhaus_drop.netset,Spamhaus - Drop"
        "https://iplists.firehol.org/files/spamhaus_edrop.netset,Spamhaus - eDrop"
        "https://www.ipdeny.com/ipblocks/data/aggregated/af-aggregated.zone,AF - Afganistan"
        "https://www.ipdeny.com/ipblocks/data/aggregated/bd-aggregated.zone,BD - Bangladesh"
        "https://www.ipdeny.com/ipblocks/data/aggregated/by-aggregated.zone,BY - Belarus"
        "https://www.ipdeny.com/ipblocks/data/aggregated/br-aggregated.zone,BR - Brazil"
        "https://www.ipdeny.com/ipblocks/data/aggregated/cn-aggregated.zone,CN - China"
        "https://www.ipdeny.com/ipblocks/data/aggregated/ir-aggregated.zone,IR - Iran"
        "https://www.ipdeny.com/ipblocks/data/aggregated/in-aggregated.zone,IN - India"
        "https://www.ipdeny.com/ipblocks/data/aggregated/kp-aggregated.zone,KP - North Korea"
        "https://www.ipdeny.com/ipblocks/data/aggregated/np-aggregated.zone,NP - Nepal"
        "https://www.ipdeny.com/ipblocks/data/aggregated/pk-aggregated.zone,PK - Pakistan"
        "https://www.ipdeny.com/ipblocks/data/aggregated/ro-aggregated.zone,RO - Romania"
        "https://www.ipdeny.com/ipblocks/data/aggregated/ru-aggregated.zone,RU - Russia"
        "https://www.ipdeny.com/ipblocks/data/aggregated/sg-aggregated.zone,SG - Singapore"
        "https://www.ipdeny.com/ipblocks/data/aggregated/sy-aggregated.zone,SY - Syria"
        "https://www.ipdeny.com/ipblocks/data/aggregated/tr-aggregated.zone,TR - Turkey"
        "https://www.ipdeny.com/ipblocks/data/aggregated/ua-aggregated.zone,UA - Ukraine"
        "https://www.ipdeny.com/ipblocks/data/aggregated/vn-aggregated.zone,VN - Vietnam"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/af-aggregated.zone,AF - Afganistan"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/bd-aggregated.zone,BD - Bangladesh"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/by-aggregated.zone,BY - Belarus"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/br-aggregated.zone,BR - Brazil"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/cn-aggregated.zone,CN - China"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/ir-aggregated.zone,IR - Iran"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/in-aggregated.zone,IN - India"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/np-aggregated.zone,NP - Nepal"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/pk-aggregated.zone,PK - Pakistan"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/ro-aggregated.zone,RO - Romania"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/ru-aggregated.zone,RU - Russia"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/sg-aggregated.zone,SG - Singapore"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/sy-aggregated.zone,SY - Syria"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/tr-aggregated.zone,TR - Turkey"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/ua-aggregated.zone,UA - Ukraine"
        "https://www.ipdeny.com/ipv6/ipaddresses/aggregated/vn-aggregated.zone,VN - Vietnam"
)

# File name with the current date and time
output_file="merged_list_${current_datetime}.txt"

# Download and merge the IP lists
for url_info in "${urls[@]}"
do
    # Splitting the URL information
    IFS=',' read -r url description <<< "$url_info"

    # Add comment with the URL and description
    echo "# URL: $url" >> "$output_file"
    echo "# Description: $description" >> "$output_file"

    echo "Download IP list from $url"
    # Download the IP list and add it to the file
    curl -sS "$url" >> "$output_file"
done

echo "Merge completed! The merged list is stored in $output_file"

# Formatting the file for the Cloudron Blocklist API
formatted_file="formatted_$output_file"

# # Add "\n" to the end of each line
awk '{printf "%s\\n",$0}' "$output_file" > "$formatted_file"

# Cloudron Blocklist API endpoint
cloudron_api_endpoint="https://yourcloudrondomain.com/api/v1/network/blocklist"

# API Key for authentication (replace 'your-api-key' with your API key)
api_key="your-api-key"

# Upload to Cloudron Blocklist API with wget
echo "# Upload to Cloudron Blocklist API with wget..."

# The data in the required format for the API
data="{\"blocklist\":\"$(cat "$formatted_file" | tr '\n' '\\n')\"}"

# Send the file with a POST request via wget
echo "$data"> temp_data.txt
wget --method=POST --header="Content-Type: application/json" --header="Authorization: Bearer $api_key" --body-file=temp_data.txt "$cloudron_api_endpoint" --quiet --output-document=output.txt

# Show the output of wget
cat output.txt

# Delete temporary files
rm temp_data.txt

[imc67]

Sorry for the Dutch language in the script, ChatGPT is wonderfully writing Dutch without asking haha.

Strange it doesn't seem to work, I extended the script later to upload it in one run to 3 different Cloudrons and that also works perfect. The part of creating the temp_data.txt is exactly the same.

• Do you see the prompt "# Upload to Cloudron Blocklist API with wget..."?
• What happens after that prompt is showed?
• It takes with these lists about 1,5 minute (on a AMD EPYC 7702P 64-Core Processor with 32GB memory) to process it via API so you have to have some patience (This is however exactly the same time like when you use the copy/paste in the GUI).
• After the script is finished the temp_data.txt is deleted, did the script ever finished?
• I execute the script in a LAMP app (on one of the Cloudrons) with only 256MB memory
• last maybe a stupid question but you've set the API token to "read/write"?

I just executed the script (my 3 Cloudrons version) en after about 5 minutes all 3 Cloudrons had the new lists with 60562 IPs (ranges) blocked.

[necrevistonnezr]

I executed the script with sh script.sh instead of bash script.sh - after doing the latter it worked!
Thanks again.

BTW the output.txt only contains {}...?

[imc67]

@necrevistonnezr said in Firewall / Spamassassin: Automatic list update:

after doing the latter it worked

aha ok, I always use ./script.sh

indeed output.txt also with me contains no data, as I read the script the right way it should contain errors during wget

[necrevistonnezr]

Super!
I have used this script now via cron and everything seems to work fine, including a significant reduction of "denied" mail attempts in the mail log.

I've added the following lines to keep the last 20 url lists, compressed with 7z (which I prefer for compression), for analysis (if needed):

7z a -mx9 "${current_datetime}.7z" "formatted_$output_file"
rm "formatted_$output_file"
rm "$output_file"
ls -td *.7z | grep -v '/$' | tail -n +20 | while IFS= read -r f; do rm -f "$f"; do>

Also: If you use the script, don't just blindly add url-lists. I already managed to lock myself out once by adding the "standard" Firehol list (https://iplists.firehol.org/files/firehol_level1.netset)

[necrevistonnezr]

With @imc67 permission, I created a community guide from this thread...

[necrevistonnezr]

... thought I got a "Failed pipeline for master | 3dbdac62" notification...

[humptydumpty]

@necrevistonnezr link to the guide, pretty please!

[necrevistonnezr]

https://docs.cloudron.io/guides/community/