# TCP-input Receiving an encoding error **URL:** \ **Category:** Logstash\ **Created:** [December 15, 2022, 8:02pm UTC](https://discuss.elastic.co/t/tcp-input-receiving-an-encoding-error/321344 "2022-12-15T20:02:29Z")\ **Posts on this page:** 3\ **Page:** 1
**Author:** ![elrozario](https://avatars.discourse-cdn.com/v4/letter/e/8e8cbc/32.png) [@elrozario](https://discuss.elastic.co/u/elrozario)\ **Post date:** [December 15, 2022, 8:02pm UTC](https://discuss.elastic.co/t/tcp-input-receiving-an-encoding-error/321344/1 "2022-12-15T20:02:29Z")
Hello, I have a TCP input and getting data with some special characters. How do I resolve this issue? Here is my config ```auto input { tcp { host => "0.0.0.0" port => "0000" mode => "server" type => "graphite-format-metrics" codec => graphite } } filter { mutate{ #gsub => ["message","\\u0000",""] #gsub => ["message", "\x86", ""] gsub => ["message", "\\u0018", ""] gsub => ["message", "¥",""] gsub => ["message", "[\[]",""] gsub => ["message", "[\]]",""] gsub => ["message", "[\\]", ""] } } output { stdout { codec => rubydebug } elasticsearch { ndex => "logstash-graphite-%{+YYYY.MM.dd}" } } ``` As you can see, I have tried to strip the unicode charecters with gsub in my filter. Getting the following so many times. ``` Received an event that has a different character encoding than you configured. {:text=\>"ㆆq\u0002a.\u0000\u0000\u0000R\x80\u0002]q\u0001U3terminal.029666.top.\_opt\_morpho\_bin\_hardmond.topcpuGA\xD8\xE6ݴ\u0000\u0000\u0000G\u0000\u0000\u0000\u0000\u0000\u0000\u0000\u0000\x86\x86q\u0002a.\u0000\u0000\u0000\x99\x80\u0002]q\u0001(U#terminal.025930.top..\_pweupl.topmemGA\xD8\xE6\xDD\xC2\xC0\u0000\u0000G@\u0000\u0000\u0000\u0000\u0000\u0000\u0000\x86\x86q\u0002U\>terminal.027151.df-mnt-configuration-lotsys.percent\_bytes-usedGA\xD8\xE6\xDD\xC3\u0000\u0000\u0000G?\xCA\u0015+?\xFF\xFF\xFD\x86\x86q\u0003e.\u0000\u0000\u0000H\x80\u0002]q\u0001U)terminal.029972.top.{chromium-bin}.topcpuGA\xD8\xE6\xDD\xC3\u0000\u0000\u0000G@\b\u0000\u0000\u0000\u0000\u0000\u0000\x86\x86q\u0002a.\u0000\u0000\u0000n\x80\u0002]q\u0001(U\eTC.RDL027702.memory.freeRamGA\xD8\xE6ݥ@\u0000\u0000GA\u0016Q\xE0\u0000\u0000\u0000\u0000\x86\x86q\u0002U\eTC.RDL027702.memory.freeRamGA\xD8\xE6ݴ@\u0000\u0000GA\u0016M\xE0\u0000\u0000\u0000\u0000\x86\x86q\u0003e.\u0000\u0000\u0000A\x80\u0002]q\u0001U\"terminal.019628.cpu.percent-activeGA\xD8\xE6\xDD\xC3\u0000\u0000\u0000G@\"UTF-8"} ```auto ``` ---
**Author:** ![Badger](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/badger/32/25190_2.png) [@Badger](https://discuss.elastic.co/u/Badger)\ **Post date:** [December 15, 2022, 8:49pm UTC](https://discuss.elastic.co/t/tcp-input-receiving-an-encoding-error/321344/2 "2022-12-15T20:49:24Z")
> [@elrozario](#): > > `gsub => ["message", "\\u0018", ""]` Your message field does not contain \u0018, it contains a UTF-16 CANCEL character. \u0018 is just the representation of that when it is printed in the log. Given how little the graphite codec [does](https://github.com/logstash-plugins/logstash-codec-graphite/blob/50a3a54c387365168d5fb15689bbc7724c50f350/lib/logstash/codecs/graphite.rb#L58), you might be better off using a line or plain codec so that you can set the encoding correctly using the charset option, and then parse the line yourself. ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [January 12, 2023, 8:50pm UTC](https://discuss.elastic.co/t/tcp-input-receiving-an-encoding-error/321344/3 "2023-01-12T20:50:06Z")
This topic was automatically closed 28 days after the last reply. New replies are no longer allowed.