The question is generic. What daemon are you using? Since you mentioned
flows, the following holds if you are using NetFlow/IPFIX and hence
nfacctd as daemon. Also the assumption is sql_dont_try_update is not set
to true (defaults to false).
Granted nothing gets lost. What you will be logged after the 900 seconds
depends on your configured aggregation method ('aggregate' directive): if
it is something like 'peer_src_ip, src_host, dst_host, proto, src_port,
dst_port, tos, timestamp_start, timestamp_end' then you will get reported
all flows during those 900 seconds (if yoo many flows, you will need to
tweak the print_cache_entries directive). If, instead, your aggregation
method is "shorter", ie. 'peer_src_ip, src_host, dst_host', then bytes
and packet counters for conversation between source/destination hosts,
yes, will be updated.
On Tue, Apr 18, 2017 at 03:14:25PM +0430, Farshid Abediny wrote:
> we set sql_refresh_time to 900 seconds due to high network list filter that
> write flows to mysql every 900 seconds so i want know when my routers send
> flows to pmacct it will save flow to its memory and every 900 seconds it
> will update database right ?
> so i do not lost flows between this interval, right?
> pmacct-discussion mailing list
pmacct-discussion mailing list