Thread: Re: curl-loader crashing network on fedora machine
Status: Alpha
Brought to you by:
coroberti
From: MLVJ <ml...@ya...> - 2008-04-01 13:06:29
|
CURL-LOADER VERSION: 0.44 HW DETAILS: CPU/S and memory are must: free total used free shared buffers cached Mem: 3109056 264216 2844840 0 13472 195048 -/+ buffers/cache: 55696 3053360 Swap: 2031608 0 2031608 grep MemTotal /proc/meminfo MemTotal: 3109056 kB grep "model name" /proc/cpuinfo model name : Intel(R) Xeon(TM) CPU 3.06GHz model name : Intel(R) Xeon(TM) CPU 3.06GHz model name : Intel(R) Xeon(TM) CPU 3.06GHz model name : Intel(R) Xeon(TM) CPU 3.06GHz HP ProLiant DL380 LINUX DISTRIBUTION and KERNEL (uname -r): uname -r 2.6.15-1.2054_FC5smp GCC VERSION (gcc -v): gcc -v Using built-in specs. Target: i386-redhat-linux Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --enable-checking=release --with-system-zlib --enable-__cxa_atexit --disable-libunwind-exceptions --enable-libgcj-multifile --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-java-awt=gtk --disable-dssi --with-java-home=/usr/lib/jvm/java-1.4.2-gcj-1.4.2.0/jre --with-cpu=generic --host=i386-redhat-linux Thread model: posix gcc version 4.1.0 20060304 (Red Hat 4.1.0-3) COMPILATION AND MAKING OPTIONS (if defaults changed): $make optimize=1 debug=0 COMMAND-LINE: ./curl-loader -f nte-conf/bulk.conf CONFIGURATION-FILE (The most common source of problems): Place the file inline here: cat bulk.conf ########### GENERAL SECTION ################################ BATCH_NAME= bulk CLIENTS_NUM_MAX=200 # Same as CLIENTS_NUM #CLIENTS_NUM_START=10 CLIENTS_RAMPUP_INC=5 INTERFACE =eth0 NETMASK=16 IP_ADDR_MIN= 10.59.250.1 IP_ADDR_MAX= 10.59.250.3 IP_SHARED_NUM=3 CYCLES_NUM= -1 URLS_NUM= 1 ########### URL SECTION #################################### URL=http://10.59.246.45/index.html #URL=http://localhost/apache2-default/ACE-INSTALL.html #URL=http://localhost/ACE-INSTALL.html URL_SHORT_NAME="local-apache" REQUEST_TYPE=GET TIMER_URL_COMPLETION = 5000 TIMER_AFTER_URL_SLEEP = 500 DOES THE PROBLEM AFFECT: COMPILATION? No LINKING? No EXECUTION? No OTHER (please specify)? Network stops. Have you run $make cleanall prior to $make ? Yes DESCRIPTION: After running the configuration file above, the network stops working - can't ping/telnet/ssh/anything to or from the machine, on the eth1 management interface. This makes it very hard to figure out what's going on, as the machine is remote to me! Note that the exact same file, but with ip addresses 192.168.1.1 and 192.168.1.3 (as per the example bulk.conf) doesn't stop the network. QUESTION/ SUGGESTION/ PATCH: How do we get it working? __________________________________________________________ Sent from Yahoo! Mail. A Smarter Inbox http://uk.docs.yahoo.com/nowyoucan.html |
From: Robert I. <cor...@gm...> - 2008-04-01 16:39:54
|
Hi, On Tue, Apr 1, 2008 at 4:06 PM, MLVJ <ml...@ya...> wrote: > CURL-LOADER VERSION: 0.44 > > HP ProLiant DL380 > > LINUX DISTRIBUTION and KERNEL (uname -r): > uname -r > 2.6.15-1.2054_FC5smp > > gcc version 4.1.0 20060304 (Red Hat 4.1.0-3) > > COMPILATION AND MAKING OPTIONS (if defaults changed): > $make optimize=1 debug=0 > > COMMAND-LINE: > > ./curl-loader -f nte-conf/bulk.conf > > CONFIGURATION-FILE (The most common source of > problems): > Place the file inline here: > > cat bulk.conf > ########### GENERAL SECTION > ################################ > > BATCH_NAME= bulk > CLIENTS_NUM_MAX=200 # Same as CLIENTS_NUM > #CLIENTS_NUM_START=10 > CLIENTS_RAMPUP_INC=5 > INTERFACE =eth0 > NETMASK=16 > IP_ADDR_MIN= 10.59.250.1 > IP_ADDR_MAX= 10.59.250.3 > IP_SHARED_NUM=3 > CYCLES_NUM= -1 > URLS_NUM= 1 > > ########### URL SECTION > #################################### > > URL=http://10.59.246.45/index.html > #URL=http://localhost/apache2-default/ACE-INSTALL.html > #URL=http://localhost/ACE-INSTALL.html > URL_SHORT_NAME="local-apache" > REQUEST_TYPE=GET > TIMER_URL_COMPLETION = 5000 > TIMER_AFTER_URL_SLEEP = 500 > > > DESCRIPTION: > After running the configuration file above, the > network stops working - can't ping/telnet/ssh/anything > to or from the machine, on the eth1 management > interface. > This makes it very hard to figure out what's going on, > as the machine is remote to me! > > Note that the exact same file, but with ip addresses > 192.168.1.1 and 192.168.1.3 (as per the example > bulk.conf) doesn't stop the network. > > QUESTION/ SUGGESTION/ PATCH: > How do we get it working? Most probably somebody is using the addresses, starting from 10, that you are trying to use. Reboot your linux (to clean up the secondary addresses from curl-loader) or delete all secondary and clean the arp caches and ping to that addresses prior to running curl-loader. Since you are the second, who reported network problems with curl-loader (the first was me, when discovered a linux kernel bug, now fixed), I suspect, that you problems may be also related to your networking card, which could be its poor or buggy support in your linux distribution. To resolve it you may wish to install FC-8 or recent Debian. Yet another option, that your routing somehow gets broken. Therefore, please, look into your routing table before running curl-loader and after. Note, that curl-loader does not effect routing tables. Run traceroute. Once I have seen a really weird case, where an old switch was auto-synched with a new server card at 10 Mbps, thus curl-loader has filled the 10 Mbps, and there was no room for any other traffic. All the above options are beyond the area of curl-loader application control. Best wishes. > > > Sincerely, > Robert Iakobashvili > "Light will come from Jerusalem" > ........................................................... > http://curl-loader.sourceforge.net > An open-source web testing and traffic generation. |