Menu
▾
▴
netacct-mysql-users — users list, ask here
You can subscribe to this list here.
| 2001 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
(3) |
Dec
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2002 |
Jan
|
Feb
(1) |
Mar
(1) |
Apr
|
May
|
Jun
|
Jul
(1) |
Aug
(5) |
Sep
(2) |
Oct
|
Nov
|
Dec
|
| 2003 |
Jan
(2) |
Feb
(1) |
Mar
|
Apr
(2) |
May
|
Jun
(1) |
Jul
(1) |
Aug
|
Sep
|
Oct
|
Nov
|
Dec
(1) |
| 2004 |
Jan
|
Feb
|
Mar
|
Apr
(1) |
May
|
Jun
|
Jul
|
Aug
|
Sep
|
Oct
|
Nov
|
Dec
|
|
From: Peter M. <Pet...@tu...> - 2004-04-14 14:42:24
|
Heyaaa!!!
i test netacct-mysql on local network, but i have problem with oversized
values:
--[ 1 ]--
i changed sql types:
ALTER TABLE `traffic` CHANGE `input` `input` BIGINT( 20 ) UNSIGNED DEFAULT '0' NOT NULL
(same thing for output, local_input, local_output, ...)
--[ 2 ]--
i tried change some "c" code in src/mysql.c:
---------------------------------------8<--------------------------------------------------------
--- mysql.c 2003-03-17 13:17:08.000000000 +0100
+++ netacct-mysql-0.76/src/mysql.c 2004-04-14 15:58:33.000000000 +0200
@@ -43,8 +43,8 @@
MYSQL_ROW r;
my_ulonglong row;
char query[8192];
- unsigned long int old_input, old_output, old_peer_input, old_peer_output;
- unsigned long int old_direct_input, old_direct_output, old_local_input, old_local_output;
+ unsigned long long int old_input, old_output, old_peer_input, old_peer_output;
+ unsigned long long int old_direct_input, old_direct_output, old_local_input, old_local_output;
int rc;
char TIME_MASK[] = "DATE_FORMAT(NOW(),'%Y-%m-%d %H:00:00')";
@@ -61,14 +61,14 @@
{
r = mysql_fetch_row(result); /* get query result */
{
- old_input = atol(r[0]);
- old_output = atol(r[1]);
- old_peer_input = atol(r[2]);
- old_peer_output = atol(r[3]);
- old_direct_input = atol(r[4]);
- old_direct_output = atol(r[5]);
- old_local_input = atol(r[6]);
- old_local_output = atol(r[7]);
+ old_input = atoll(r[0]);
+ old_output = atoll(r[1]);
+ old_peer_input = atoll(r[2]);
+ old_peer_output = atoll(r[3]);
+ old_direct_input = atoll(r[4]);
+ old_direct_output = atoll(r[5]);
+ old_local_input = atoll(r[6]);
+ old_local_output = atoll(r[7]);
}
old_input+=(tmpData->nPeerFlag==0)?tmpData->nInTrafic:0;
@@ -81,11 +81,11 @@
old_local_output+=(tmpData->nPeerFlag==3)?tmpData->nOutTrafic:0;
- /* syslog(LOG_INFO,"last record is - %li",old_trafic); */
- /* syslog(LOG_INFO,"sum = %li",old_trafic); */
+ /* syslog(LOG_INFO,"last record is - %Lu",old_trafic); */
+ /* syslog(LOG_INFO,"sum = %Lu",old_trafic); */
DEBUG(DBG_STATE,sprintf(dbg,"upate info for ip %s, peer_flag=%u\n",spyip,tmpData->nPeerFlag));
- sprintf(query,"UPDATE %s SET input='%li', output='%li', peer_input='%li', peer_output='%li', direct_input='%li', direct_output='%li', local_input='%li',local_output='%li' WHERE ip='%s' AND time=%s", "traffic", old_input,old_output, old_peer_input, old_peer_output,old_direct_input, old_direct_output,old_local_input,old_local_output, spyip,TIME_MASK);
+ sprintf(query,"UPDATE %s SET input='%Lu', output='%Lu', peer_input='%Lu', peer_output='%Lu', direct_input='%Lu', direct_output='%Lu', local_input='%Lu',local_output='%Lu' WHERE ip='%s' AND time=%s", "traffic", old_input,old_output, old_peer_input, old_peer_output,old_direct_input, old_direct_output,old_local_input,old_local_output, spyip,TIME_MASK);
@@ -104,7 +104,7 @@
DEBUG(DBG_STATE,sprintf(dbg,"upate info for ip %s\n",spyip));
- sprintf(query,"INSERT INTO %s (ip,time, input,output,peer_input,peer_output,direct_input,direct_output,local_input,local_output) VALUES ( '%s',%s,'%li','%li','%li','%li','%li','%li','%li','%li')", "traffic", spyip, TIME_MASK, old_input,old_output,old_peer_input,old_peer_output,old_direct_input,old_direct_output, old_local_input,old_local_output);
+ sprintf(query,"INSERT INTO %s (ip,time, input,output,peer_input,peer_output,direct_input,direct_output,local_input,local_output) VALUES ( '%s',%s,'%Lu','%Lu','%Lu','%Lu','%Lu','%Lu','%Lu','%Lu')", "traffic", spyip, TIME_MASK, old_input,old_output,old_peer_input,old_peer_output,old_direct_input,old_direct_output, old_local_input,old_local_output);
@@ -143,7 +143,7 @@
return 0;
}
-// syslog(LOG_DEBUG,"ip = %s, In = %li, Out = %li, peer = %i", intoa(tmpData->ipAddress), tmpData->nInTrafic, tmpData->nOutTrafic, tmpData->nPeerFlag);
+// syslog(LOG_DEBUG,"ip = %s, In = %Lu, Out = %Lu, peer = %i", intoa(tmpData->ipAddress), tmpData->nInTrafic, tmpData->nOutTrafic, tmpData->nPeerFlag);
/* and write it to mysql*/
if((tmpData->nInTrafic != 0) || (tmpData->nOutTrafic != 0))
write_mysql(tmpData, mysql);
@@ -152,7 +152,7 @@
while(tmpData = (struct HOST_DATA*) GetNextHostData())
{
-// syslog(LOG_DEBUG,"ip = %s, In = %li, Out = %li, peer = %i", intoa(tmpData->ipAddress), tmpData->nInTrafic, tmpData->nOutTrafic, tmpData->nPeerFlag);
+// syslog(LOG_DEBUG,"ip = %s, In = %Lu, Out = %Lu, peer = %i", intoa(tmpData->ipAddress), tmpData->nInTrafic, tmpData->nOutTrafic, tmpData->nPeerFlag);
i++;
/* and write data to mysql
* if IN and OUT traffic are not NULL
---------------------------------------8<--------------------------------------------------------
but there are still some mistakes ... my changes are insufficient ;-((
help me, please! thanks
--
5o Peter.Mann at tuke.sk
KLFMANiK ICQ 12491471
PM2185-RIPE
|
|
From: Egor <kn...@me...> - 2003-12-20 14:37:04
|
Hello netacct-mysql-users, Does netacct-mysql compiles on FreeBSD 5.1 ?? 'make' and 'gmake' get errors on /usr/include/malloc.h What can I do to fix this? -- Best regards, Egor mailto:kn...@me... |
|
From: Alex <ale...@rd...> - 2003-07-01 14:59:44
|
I'm using netacct-mysql-0.76 and netstat-3.04_1 and it seems that it give the wrong count on the number of megabytes passed in/out . Maybe anyone can help me with this.../me very puzzled and clueless. Here is my config: rh9 with kernel 2.4.20 which holds a private network with NAT. I have also tried the same config without NAT, only with squid setup, without any kind of NAT. I did some tests by downloading trough squid a 50 megabytes file and some browsing for 30 minutes and nestat only display something like 2,3 or 4 megabytes...something similar...anyway nothing not even close to the real amount of traffic (not counting the traffic generated by the rest of the users on the privaet network ..and they did work on the net at that time). Thanks in advance for any hints on this! Alex --------------------------------------- My files: /usr/local/etc/naccttab : sniff 0 database mysql mysql_user acct mysql_password xxxxx mysql_host 192.168.254.1 mysql_port 3306 mysql_database netacct pidfile /var/run/nacctd.pid compactnet 192.168.254.0 255.255.255.0 flush 330 fdelay 70 device eth1 device eth2 ignorenet 127.0.0.0 255.0.0.0 debug 0 debugfile /tmp/nacctd.debug headers tr 40 38 headers lo 14 12 headers eth 14 12 |
|
From: <ga...@rc...> - 2003-06-02 10:07:13
|
Hello! I downloaded NetStat and installed it. It works but there is a directory netstat-mrtg with netstat-mrtg , netacct.html and sample.mrtg.cfg files without any documentation for usage! In netacct-mysql's contrib directory is also a file mrta2netacct without any documentation! Can somebody help me in using those files? Thanks |
|
From: Nikolay H. <ni...@st...> - 2003-04-10 07:08:50
|
are you correctly entered compactnet net netmask ? and can you connect with mysql -u acct -p with correct password to mysql? Nikolay Hristov ----- Original Message ----- From: "Szemerédy Gábor" <ga...@rc...> To: <net...@li...> Sent: Tuesday, April 08, 2003 12:16 PM Subject: [Netacct-mysql-users] Data collection > Hello! > I installed the product on redhat 6.2 > The installation was ok > The daemon (nacctd) is running , mysql is running but no data appears > in the traffic table > There is plenty of empty files in /tmp directory except netacctd.debug > which is not empty > How can I see if nacctd is collecting data or not? > Thanks |
|
From: <ga...@rc...> - 2003-04-08 09:16:42
|
Hello! I installed the product on redhat 6.2 The installation was ok The daemon (nacctd) is running , mysql is running but no data appears in the traffic table There is plenty of empty files in /tmp directory except netacctd.debug which is not empty How can I see if nacctd is collecting data or not? Thanks |
|
From: Brian I. <mon...@ne...> - 2003-02-01 08:04:52
|
Hi! Excuse me for saying so, but I don't think the instructions for configuration in the naccttab file are very specific. I have a host with a couple of networks cards, that gathers traffic from 3 subnets, 192.168.0.0/24, 192.168.1.0/24 and 192.168.3.0/24. These subnets are local, and I've added them as "compactnet" What makes me wonder is the "ournet" and "direct_peer" stuff. The 3 subnets above all goes through a firewall with NAT, so my own subnet addresses won't be seen by the sniffer. The funny stuff is that all trffic between hosts on the 192.168.1.0 network is displayed as "international" in the netstat web-interface - instead of local... What am I doing wrong ?? Regards, /Brian |
|
From: Jeremy F. <net...@50...> - 2003-01-15 16:45:32
|
Hi, Does anyone who is using netacct know of a program that will read the nacctd.debug file and show real time stat reports? We are currently using the old netacct-mysql-0.73rc4 version. Does the new version have the ability to show real time (or close to real time) usage stats? Thanks in advance, Jeremy Fox |
|
From: Georgi A. <g.a...@di...> - 2003-01-05 13:04:29
|
Here is Apache error_log: [Sun Jan 5 14:57:58 2003] [error] PHP Notice: Undefined variable: = action in /var/www/htdocs/stat/index.php on line 49 [Sun Jan 5 14:57:58 2003] [error] PHP Notice: Undefined variable: mtr = in /var/www/htdocs/stat/index.php on line 62 |
|
From: Nikolay H. <ge...@st...> - 2002-09-27 18:18:58
|
hello to all subscribers :) in a few days i'll release new version of netacct-mysql. here are = important changes: - full logging is no longer available .. if you want to log all = conncetions in sql go and get 0.73 it will work just fine. accounting = table removed so if you upgrade and don't need this data you can safely = remove this table. - listen to more that 1 interface, some code rewritten and now using = pthreads .. every interface get its own thread - removed some code about writing in file - now it logs current traffic not only when connection is closed .. = there are no net-acct-dump.o.x.x files any more - support of 4 types traffic - international/peering/direct/local .. so = you can differ traffic in 4 stages - php upgrade script which will convert old traffic in new type = (0.73->0.75) .. 0.74rc1 -> 0.75 don't need any upgrade some explanation: 1. the reason to remove full logging is that holding every connection in = memory and in five minutes write it in sql causes heavy load of mysql in = case you have a heavy network load ... it leads to "mysql resource = temporary unavailable" when strace-ing forked process which writes data = in mysql and possibly lost of accounting data/pc crash/mysql 99% cpu. old way collecting data: you open a web page and then open an ftp and get a file: your_ip web_server_ip send_traffic peer_flag web_server_ip your_ip received_traffic peer_flag your_ip ftp_server_ip send_traffic peer_flag ftp_server_ip your_ip received_traffic peer_flag new way collecting data: your_ip in_traffic out_traffic peer_flag as you see this will not consume large amounts of memory ... there will = be ip's_in_compactnet*peer_types*13bytes in memroy (peer_types =3D 5) 2. i've read some messages from groups.google.com about problems with = libpcap+pthreads .. i've read some code and finally got it to work but = it is tested inly on linux so please go and get latest cvs version and = try it on some bsd .. some ideas, code and #ifdef's taken from = snort/ntop/ethereal ..=20 i;m running this about 3 days and no problems at all on 10Mbit LAN with = > 300 pc's and heavy load of traffic so i'll test it some more time and = will release final version Regards, Nikolay Hristov |
|
From: Hilko B. <be...@vd...> - 2002-09-23 11:31:22
|
ulog-acctd is a userspace network accounting daemon which generates log files of network traffic for accounting purposes. Its output format cen be configured via a format string, therefore it's possible to have it look similar to that of net-acct. I had originally hacked net-acctd for my employer (an ISP that uses Linux for some core routers) to use the Linux 2.2 packet interface (packet(7)) instead of the now obsolete Linux 2.0 interface (ip(7)), but we found out that a significant number of packets were not accounted under high network load, so we abandoned that approach. We suspect that the "old" net-acct (and also net-acct-mysql) share this problem. As the name suggests, ulog-acctd uses the ULOG netfilter target that is provided with newer Linux 2.4 kernels. This means that ulog-acctd can only see those packets that are handled by netfilter. Packets that merely travel through the same ethernet segment are not handled. Because selection of packets that are to be accounted can be done in kernel space via iptables(8), ulog-acctd does not need code for that and thus does not do much more than the actual aggregation and logging of accounting information. You can get ulog-acctd from http://savannah.nongnu.org/download/ulog-acctd/. -Hilko |
|
From: Lorenz B. <lb...@pr...> - 2002-08-25 17:04:14
|
SGVsbG8sDQogDQpub3AsIHlvdSBhcmUgbm90IHRoZSBvbmx5IG9uZSA6LSkgSW0gb24gdGhlIExp c3QgYXMgd2VsbCA6LSkNCiANCkJ5ZSBMb3JlbnoNCg0KCS0tLS0tVXJzcHLDvG5nbGljaGUgTmFj aHJpY2h0LS0tLS0gDQoJVm9uOiBGcmllZHJpY2ggTG9iZW5zdG9jayBbbWFpbHRvOmZsQGZsLnBy aXYuYXRdIA0KCUdlc2VuZGV0OiBTbyAyNS4wOC4yMDAyIDEyOjE3IA0KCUFuOiBOZXRhY2N0LU15 U1FMIE1haWxpbmcgTGlzdGUgDQoJQ2M6IA0KCUJldHJlZmY6IFtOZXRhY2N0LW15c3FsLXVzZXJz XSBBTlkgZGV2ZWxvcGVycyBvbiB0aGUgbGlzdCA/Pz8/Pw0KCQ0KCQ0KDQoJSGkhDQoJDQoJQVJF IHRoZXJlIEFOWSBkZXZlbG9wZXJzIG9uIHRoZSBsaXN0PyBJdCBzZWVtcyB0aGF0IEknbSB0aGUg b25seSBwZXJzb24NCglwb3N0aW5nIHRvIHRoaXMgbGlzdC4NCgkNCgktLQ0KCU1mRyAvIFJlZ2Fy ZHMNCglGcmllZHJpY2ggTG9iZW5zdG9jaw0KCV9fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fDQoJRnJpZWRyaWNoIExvYmVu c3RvY2sgICAgICAgICBGTDIyNi1SSVBFICAgICAgICAgICAgIEludGVybmV0c2VydmljZXMNCglV Ukw6IGh0dHA6Ly93d3cuZmwucHJpdi5hdC8gICAgICAgICAgICAgICAgICAgICBFbWFpbDogZmxA ZmwucHJpdi5hdA0KCV9fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fDQoJDQoJDQoJDQoJLS0tLS0tLS0tLS0tLS0tLS0tLS0t LS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLS0tLQ0KCVRoaXMgc2YubmV0IGVtYWlsIGlz IHNwb25zb3JlZCBieTogT1NETiAtIFRpcmVkIG9mIHRoYXQgc2FtZSBvbGQNCgljZWxsIHBob25l PyAgR2V0IGEgbmV3IGhlcmUgZm9yIEZSRUUhDQoJaHR0cHM6Ly93d3cuaW5waG9uaWMuY29tL3Iu YXNwP3I9c291cmNlZm9yZ2UxJnJlZmNvZGUxPXZzMzM5MA0KCV9fX19fX19fX19fX19fX19fX19f X19fX19fX19fX19fX19fX19fX19fX19fX19fDQoJTmV0YWNjdC1teXNxbC11c2VycyBtYWlsaW5n IGxpc3QNCglOZXRhY2N0LW15c3FsLXVzZXJzQGxpc3RzLnNvdXJjZWZvcmdlLm5ldA0KCWh0dHBz Oi8vbGlzdHMuc291cmNlZm9yZ2UubmV0L2xpc3RzL2xpc3RpbmZvL25ldGFjY3QtbXlzcWwtdXNl cnMNCgkNCg0K |
|
From: Friedrich L. <fl...@fl...> - 2002-08-25 10:19:42
|
Hi! ARE there ANY developers on the list? It seems that I'm the only person posting to this list. -- MfG / Regards Friedrich Lobenstock ____________________________________________________________________ Friedrich Lobenstock FL226-RIPE Internetservices URL: http://www.fl.priv.at/ Email: fl...@fl... ____________________________________________________________________ |
|
From: Friedrich L. <fl...@fl...> - 2002-08-22 21:09:39
|
Friedrich Lobenstock wrote: > Hi! > > I see that some time that when nacctd forks in process.c for processing > the SQL selects and updates it takes up a lot of memory - steadily > increasing > till it is somehow able to send all data. nacctd currently at 1GB !! 11:05pm up 36 days, 12:23, 1 user, load average: 1.77, 2.09, 2.02 81 processes: 76 sleeping, 5 running, 0 zombie, 0 stopped CPU states: 73.9% user, 26.0% system, 0.0% nice, 0.0% idle Mem: 255456K av, 251060K used, 4396K free, 0K shrd, 14972K buff Swap: 1052632K av, 935196K used, 117436K free 15132K cached PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND 1197 root 20 0 1038M 159M 8020 S 0.0 63.9 0:28 nacctd 11056 root 20 0 19604 19M 88 R 3.7 7.6 28:40 nacctd 26064 root 20 0 4264 2260 1808 S 0.0 0.8 0:03 xconsole 16962 root 20 0 2208 2164 1696 R 0.0 0.8 0:00 sshd 17331 root 20 0 1796 1796 1600 S 0.0 0.7 0:44 ntpd 5217 named 20 0 2512 1680 780 S 0.0 0.6 0:11 named 16992 root 20 0 1580 1580 1168 S 0.0 0.6 0:00 bash 9987 mysql 20 0 7896 1088 836 S 0.0 0.4 0:00 mysqld-max 9989 mysql 20 0 7896 1088 836 S 0.0 0.4 0:00 mysqld-max 9990 mysql 20 0 7896 1088 836 S 0.0 0.4 0:00 mysqld-max 9991 mysql 20 0 7896 1088 836 S 0.0 0.4 0:00 mysqld-max 1201 mysql 0 0 7896 1088 836 R 84.7 0.4 423:52 mysqld-max ... -- MfG / Regards Friedrich Lobenstock ____________________________________________________________________ Friedrich Lobenstock FL226-RIPE Internetservices URL: http://www.fl.priv.at/ Email: fl...@fl... ____________________________________________________________________ |
|
From: Friedrich L. <fl...@fl...> - 2002-08-21 20:29:07
|
Hi! I see that some time that when nacctd forks in process.c for processing the SQL selects and updates it takes up a lot of memory - steadily increasing till it is somehow able to send all data. See this output of "top": 10:17pm up 35 days, 11:35, 1 user, load average: 2.04, 2.19, 2.15 81 processes: 76 sleeping, 5 running, 0 zombie, 0 stopped CPU states: 63.5% user, 30.0% system, 6.3% nice, 0.0% idle Mem: 255456K av, 251320K used, 4136K free, 0K shrd, 21152K buff Swap: 1052632K av, 200044K used, 852588K free 16544K cached PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND 29245 root 20 0 329M 162M 4292 S 0.0 65.2 0:09 nacctd 11056 root 20 0 6168 6076 88 S 1.1 2.3 3:31 nacctd 5408 wwwrun 20 0 4680 2796 2296 S 0.0 1.0 0:04 httpd 5399 wwwrun 20 0 5232 2792 2272 S 0.0 1.0 0:03 httpd 9987 mysql 20 0 7356 2352 1856 S 0.0 0.9 0:00 mysqld-max 9989 mysql 20 0 7356 2352 1856 S 0.0 0.9 0:00 mysqld-max 9990 mysql 20 0 7356 2352 1856 S 0.0 0.9 0:00 mysqld-max 9991 mysql 20 0 7356 2352 1856 S 0.0 0.9 0:00 mysqld-max 29252 mysql 0 0 7356 2352 1856 R 75.6 0.9 127:12 mysqld-max 26064 root 20 0 3952 2108 1808 S 0.0 0.8 0:03 xconsole 5217 named 20 0 2508 1868 808 S 0.0 0.7 0:05 named 17331 root 20 0 1796 1796 1600 S 0.0 0.7 0:44 ntpd # ps fwxa 11056 ? R 3:38 /usr/local/sbin/nacctd 29245 ? S 0:09 \_ /usr/local/sbin/nacctd Can someone have a look where the code might consume more and more memory and not releasing it? Maybe the mySQL results are not freed before each new request or the like ?! -- MfG / Regards Friedrich Lobenstock ____________________________________________________________________ Friedrich Lobenstock FL226-RIPE Internetservices URL: http://www.fl.priv.at/ Email: fl...@fl... ____________________________________________________________________ |
|
From: Friedrich L. <fl...@fl...> - 2002-08-19 22:43:40
|
Hi! As usuall I was to early, just found nacctd consuming about 250MB and counting memory and mysql daemon at nearly 100%. Some minutes later it was down to about 20MB. Restarting the daemon and it is down to about 800KB. What can make nacctd react so badly? -- MfG / Regards Friedrich Lobenstock -------- Original Message -------- Subject: [ netacct-mysql-Bugs-597108 ] [0.74r1] MySQL-Daemon goes to 100% Date: Mon, 19 Aug 2002 09:00:59 -0700 From: no...@so... To: no...@so... Bugs item #597108, was opened at 2002-08-19 13:59 You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=378119&aid=597108&group_id=23298 Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Friedrich Lobenstock (friedl) Assigned to: Nobody/Anonymous (nobody) Summary: [0.74r1] MySQL-Daemon goes to 100% Initial Comment: With 0.74r1 the mysql daemon goes to 100% after a while and nacctd builds up memory. With 0.73 this problem does not exist, at least not up to now ;-) ---------------------------------------------------------------------- >Comment By: Friedrich Lobenstock (friedl) Date: 2002-08-19 18:00 Message: Logged In: YES user_id=38090 Seems to be a problem because the default value for flush is 300. Now I'm down to 30 seconds and the situations seems to be more stable. PS: I'm using the follwing config: database mysql mysql_user acct mysql_password XXXXXXX mysql_host localhost mysql_port 0 mysql_database netacct mysql_table accounting pidfile /var/run/nacctd.pid compactnet AAA.BBB.CCC.0 255.255.255.0 ignoremask 255.255.255.0 flush 30 fdelay 60 file /var/log/net-acct dumpfile /var/log/net-acct-dump notdev eth0 device eth1 ignoremask 255.255.255.0 ignorenet 127.0.0.0 255.0.0.0 debug 2 debugfile /var/log/nacctd.debug headers tr 40 38 headers lo 14 12 headers isdn 4 0 headers eth 14 12 headers plip 14 12 ---------------------------------------------------------------------- Comment By: Friedrich Lobenstock (friedl) Date: 2002-08-19 16:44 Message: Logged In: YES user_id=38090 Hmmmm....a downgrade did not really help. nacctd starts building up memory and mysql is using a much cpu as it can get. Output from top: PID USR PRI NI SIZE RSS SHARE ST %CPU %MEM TIME COMMAND 26585 root 20 0 51884 50M 764 S 0.0 20.3 0:01 nacctd 14029 wwwrun 20 0 7876 7496 5312 S 0.0 2.9 0:04 httpd 1884 mysql 20 0 5444 5444 2320 S 0.0 2.1 0:00 mysqld-max 1886 mysql 20 0 5444 5444 2320 S 0.0 2.1 0:00 mysqld-max 1887 mysql 20 0 5444 5444 2320 S 0.0 2.1 0:00 mysqld-max 1889 mysql 20 0 5444 5444 2320 S 0.0 2.1 0:00 mysqld-max 14352 wwwrun 20 0 6360 5444 3548 S 0.0 2.1 0:04 httpd 26586 mysql 0 0 5444 5444 2320 R 97.3 2.1 15:55 mysqld-max 14011 root 20 0 4988 4228 3824 S 0.0 1.6 0:00 httpd 26064 root 20 0 3712 3576 916 S 0.0 1.3 0:02 xconsole 27074 root 20 0 2216 2172 1692 R 0.9 0.8 0:01 sshd 30985 root 0 0 2412 2012 1668 S 0.0 0.7 0:47 snmpd 17331 root 20 0 1796 1796 1600 S 0.0 0.7 0:44 ntpd 27118 root 20 0 1712 1712 1252 S 0.0 0.6 0:00 bash ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=378119&aid=597108&group_id=23298 |
|
From: Clemens R. <cre...@gm...> - 2002-07-18 17:55:38
|
Hi,
I am currently experiencing problems with the compactnet mode of
netacct-mysql-0.73
Here is the config:
---
compactnet 192.168.0.0 255.255.255.0
flush 300
fdelay 60
device eth0
iflimit eth0
ignoremask 255.255.255.0
ignorenet 127.0.0.0 255.0.0.0
---
the data is collected as expected, but the system load is up to 3,0.
To see what nacctd does, I started
mysqld_save --log=mysql.log
After starting nacctd, there is no sql UPDATE in table traffic for 5 minutes.
But then every packet seems to be logged because loads of UPDATE
messages are logged for the same IP address.
Is compact mode not working correctly or am I doing something wrong?
Best regards...
- Clemens -
|
|
From: Nikolay H. <ge...@st...> - 2002-03-02 13:23:52
|
> Message: 1 > Date: Tue, 26 Feb 2002 11:24:00 +0200 > From: Stoilis Giannis <st...@th...> > To: net...@li... > Subject: [Netacct-mysql-users] no detailled logging... > > Hello, I have a problem with netacct NOT logging the traffic after > working perfectly for 2 days. My configuration: > Ethernet Card: 3COM 905C-TX > 3c59x: Donald Becker and others. www.scyld.com/network/vortex.html > 02:09.0: 3Com PCI 3c905C Tornado at 0x1400. Vers LK1.1.16 > > My networks consists of several subnets(10.0.0.x, 10.0.1.x and so on) > > OS: Suse Linux 7.2, custom kernel 2.4.16 > > A router in my network forwards all traffic (0.0.0.0) to this server. > This server does transparent proxying with squid, so ANY traffic > destined to the internet passes right through it. > Then, it forwards any request to my border router with the serial link > to the internet. i've not tested nacctd with transparent proxying but it should work .. > > I am using netacct-mysql-0.73rc4 with MySQL 3.23.37. > > More details about the problem: > summary traffic is updated correctly.(traffic table). The accounting > table is not updated at all or is updated with strange lines like: > 84 10:52:41 2002-02-26 0 17 0.0.0.0 68 255.255.255.255 67 2 676 unknown > eth0 0 i think these strange addresses are from windows machines which are doing some broadcast requests > I can see that the dump files has the correct entries in it when I view > it. I can assume that it flushes it correctly since after grows to some > kilobytes and then it empties it. > I have raised the debug level but I could not understand exactly what to > look for in the debug file. dump file is for emergency purpiose .. i mean if your pc crashes .. all data stays in dump file .. if everything is ok .. every time when it writes to mysql it empties this file. if you have files with .o .o.o .o.o.o extensions it means that this data is noy written in mysql .. there is a script in contrib dir which you can use to import this data in mysql > > what is more strange, is that is worked for 2 days and then just > stopped... I even tryied completely erasing the configuration file and > then just using the default one, changing only the subnet ip it was > supposed to monitor. it is very strange thing .. i see such thing for the first time .. i assume that there is no problems with mysql server? > here it is: (/usr/local/etc/naccttab) > > database mysql > mysql_user acct > mysql_password acct_password > mysql_host localhost > mysql_database netacct > mysql_table accounting > pidfile /var/run/nacctd.pid > compactnet 10.0.0.0 255.0.0.0 > flush 1 > fdelay 60 > file /var/log/net-acct > dumpfile /var/log/net-acct-dump > device eth0 > ignoremask 255.0.0.0 remove this line (ignoremask 255.0.0.0) > ignorenet 127.0.0.0 255.0.0.0 > debug 2 > debugfile /tmp/nacctd.debug > headers tr 40 38 > headers lo 14 12 > headers isdn 4 0 > headers eth 14 12 > headers plip 14 12 > > Any ideas? Any suggestion would be most helpful... > Thanks in advance. > > - Stoilis Giannis > Nikolay Hristov |
|
From: Stoilis G. <st...@th...> - 2002-02-26 09:24:11
|
Hello, I have a problem with netacct NOT logging the traffic after working perfectly for 2 days. My configuration: Ethernet Card: 3COM 905C-TX 3c59x: Donald Becker and others. www.scyld.com/network/vortex.html 02:09.0: 3Com PCI 3c905C Tornado at 0x1400. Vers LK1.1.16 My networks consists of several subnets(10.0.0.x, 10.0.1.x and so on) OS: Suse Linux 7.2, custom kernel 2.4.16 A router in my network forwards all traffic (0.0.0.0) to this server. This server does transparent proxying with squid, so ANY traffic destined to the internet passes right through it. Then, it forwards any request to my border router with the serial link to the internet. I am using netacct-mysql-0.73rc4 with MySQL 3.23.37. More details about the problem: summary traffic is updated correctly.(traffic table). The accounting table is not updated at all or is updated with strange lines like: 84 10:52:41 2002-02-26 0 17 0.0.0.0 68 255.255.255.255 67 2 676 unknown eth0 0 I can see that the dump files has the correct entries in it when I view it. I can assume that it flushes it correctly since after grows to some kilobytes and then it empties it. I have raised the debug level but I could not understand exactly what to look for in the debug file. what is more strange, is that is worked for 2 days and then just stopped... I even tryied completely erasing the configuration file and then just using the default one, changing only the subnet ip it was supposed to monitor. here it is: (/usr/local/etc/naccttab) database mysql mysql_user acct mysql_password acct_password mysql_host localhost mysql_database netacct mysql_table accounting pidfile /var/run/nacctd.pid compactnet 10.0.0.0 255.0.0.0 flush 1 fdelay 60 file /var/log/net-acct dumpfile /var/log/net-acct-dump device eth0 ignoremask 255.0.0.0 ignorenet 127.0.0.0 255.0.0.0 debug 2 debugfile /tmp/nacctd.debug headers tr 40 38 headers lo 14 12 headers isdn 4 0 headers eth 14 12 headers plip 14 12 Any ideas? Any suggestion would be most helpful... Thanks in advance. - Stoilis Giannis |
|
From: Nikolay H. <ge...@st...> - 2001-11-07 21:04:34
|
* a bug which cause incorrect port number logging * missing documentation about upgrading of mysql table * annoying bug uncommented debug string which fills syslog with mysql queries Nikolay Hristov |
|
From: Nikolay H. <ge...@st...> - 2001-11-06 22:20:53
|
this is test |
|
From: Nikolay H. <ge...@st...> - 2001-11-06 18:48:25
|
Helpthis is test |
