We currently run PureMessage using a locally clustered+mirrored SQL Server installation with an SQL server instance on each of the mailbox store hosts.
Any messages that enter the organization or leave the Exchange servers are additionally scanned on the mail relay servers, so PureMessage effectively only needs to be present to handle infected traffic between local Exchange users (happens almost never).
Since that installation was planned, however, other parts of the organization have grown a centrally administered SQL server installation. Hence we'd be happy to use an external SQL server instead of managing our own SQL server infrastructure locally on the Exchange servers.
However, the QoS guarantee for the central SQL server includes something like a "maybe inaccessible for a short time (few minutes) during cluster failover" disclaimer.
Given the operating range, we think we can actually live with "during this time, PureMessage does not scan any messages and operates in pass-through mode".
However, I cannot find any resources on this topic.
How does PureMessage react, when the SQL database connection is lost?
- Does mail processing go on hold (not really an option)?
- Do all mails transmitted during the SQL server downtime get marked as "unscanned" (acceptable)?
- Does mail processing continue without virus scanning (bad security-wise, but more or less what we want)?
Perhaps someone who had this happen as part of an (unplanned) outtage can comment on this.
This thread was automatically locked due to age.