How to retrieve a string that starts with another string using grok?

[Siri_sour]

Hi everyone,
I just starting working with es and kibana, right now I'm trying to structure this log :

2019-08-13 03:37:49,738 [default task-28] ERROR [org.jboss.jca.core.connectionmanager.listener.TxConnectionListener] sessionhash="8263c56211a7cc77fdd047d09b6ab8d2" requestid="56510X1565681869629" IJ000315: Pool QueueConnectionFactory has 1 active handles

I wan't to get something like this:
{
"timestamp": "2019-08-13 03:37:49,738"
"task": "default task-28",
"loglevel": "ERROR",
"package": "org.jboss.jca.core.connectionmanager.listener.TxConnectionListener",
"sessionhash":"8263c56211a7cc77fdd047d09b6ab8d2",
"requestid"="56510X1565681869629",
"message"="IJ000315: Pool QueueConnectionFactory has 1 active handles"

}

How can I retrieve the sessionhash and requestid ?

Thanks in advance !

[Badger]

The following would do it

    dissect { mapping => { "message" => "%{[@metadata][timestamp]} %{+[@metadata][timestamp]} [%{task}] %{loglevel} [%{package}] %{[@metadata][restOfLine]}" } }
    date { match => [ "[@metadata][timestamp]", "YYYY-MM-dd HH:mm:ss,SSS" ] }
    grok {
        match => {
            "[@metadata][restOfLine]" => [
                '^sessionhash="%{BASE16NUM:sessionhash}" requestid="%{WORD:requestid}" %{GREEDYDATA:message}'
            ]
        }
        overwrite => [ "message" ]
    }

I use dissect to parse the first part of the message which will be common to different messages. (I am assuming you later want to parse other messages, if that is not true you can fold the grok into the dissect and do it all in one filter.)

[Siri_sour]

Thank you so much, I did some modifications but this was the exact thing I needed.