Fazer uma PerguntaFazer uma Pergunta
 

RespondidoAntigen best practices

  • quarta-feira, 3 de junho de 2009 8:48ief Medalhas de usuárioMedalhas de usuárioMedalhas de usuárioMedalhas de usuárioMedalhas de usuário
     
    Hello,

    We use Antigen 9.1 1097 with the default installation. What are the Best Practices to tune Antigen (how many scanners, BIAS settings etc.) . We see for example 100% CPU on Antigen services and got "messages exceed internet timeout".

    Thanks

Respostas

  • terça-feira, 9 de junho de 2009 10:43Andy S. Day Medalhas de usuárioMedalhas de usuárioMedalhas de usuárioMedalhas de usuárioMedalhas de usuário
     Respondido
    Hi IEF,

    Our Best Practices Guide should give you some good pointers. In particular, look at the following settings:
       o # engines enabled (per scanjob) - even reducing this to 3 or 4 gives you a high level of protection;
       o Engine Bias - I would advise setting this to 'Max Certainty', 'Favor Certainty', or 'Neutral', in conjunction with 3 or 4 enabled engines. 'Neutral' will be the most efficient of the 3 settings;
       o 'Realtime Process Count' and 'Internet Process Count' - these are found under SETTINGS>General Options and are set to 2 by default. If you have previously raised these any higher, I would consider moving them back to 2. This gives Antigen 2 channels to talk to Exchange with / to scan messages, per Storage Group and for the SMTP scanjob;
       o If you are using RBL servers and can determine that lookups are taking a long time to come back to Antigen, consider disabling 1 or more of these;
       o If 'Perform Reverse DNS Lookup' (under SETTINGS>General Options) is enabled at all, consider disabling it ('Disable All'), again if you can determine that lookups are taking a long time to come back to Antigen.

    If you feel you have optimised your settings and are still receiving timeouts while scanning messages, I would advise you to open a Support Case with CSS (Customer Service and Support).

    Kind Regards,

    Andy Day

    CSS Security, Sr. Support Engineer (Antigen/Forefront Server Security)

Todas as Respostas

  • terça-feira, 9 de junho de 2009 10:43Andy S. Day Medalhas de usuárioMedalhas de usuárioMedalhas de usuárioMedalhas de usuárioMedalhas de usuário
     Respondido
    Hi IEF,

    Our Best Practices Guide should give you some good pointers. In particular, look at the following settings:
       o # engines enabled (per scanjob) - even reducing this to 3 or 4 gives you a high level of protection;
       o Engine Bias - I would advise setting this to 'Max Certainty', 'Favor Certainty', or 'Neutral', in conjunction with 3 or 4 enabled engines. 'Neutral' will be the most efficient of the 3 settings;
       o 'Realtime Process Count' and 'Internet Process Count' - these are found under SETTINGS>General Options and are set to 2 by default. If you have previously raised these any higher, I would consider moving them back to 2. This gives Antigen 2 channels to talk to Exchange with / to scan messages, per Storage Group and for the SMTP scanjob;
       o If you are using RBL servers and can determine that lookups are taking a long time to come back to Antigen, consider disabling 1 or more of these;
       o If 'Perform Reverse DNS Lookup' (under SETTINGS>General Options) is enabled at all, consider disabling it ('Disable All'), again if you can determine that lookups are taking a long time to come back to Antigen.

    If you feel you have optimised your settings and are still receiving timeouts while scanning messages, I would advise you to open a Support Case with CSS (Customer Service and Support).

    Kind Regards,

    Andy Day

    CSS Security, Sr. Support Engineer (Antigen/Forefront Server Security)
  • quinta-feira, 11 de junho de 2009 8:02ief Medalhas de usuárioMedalhas de usuárioMedalhas de usuárioMedalhas de usuárioMedalhas de usuário
     
    Thanks for the information