Dwww list of documentation (long)

Wayback MachineAbout this captureCOLLECTED BY Organization: Alexa Crawls Starting in 1996, Alexa Internet has been donating their crawl data to the Internet Archive. Flowing in every day, these data are added to the Wayback Machine after an embargo period. Collection: Alexa Crawls DE Crawl data donated by Alexa Internet. This data is currently not publicly accessible TIMESTAMPSloading

To the short index

The list of documents is divided into sections:

[Bridge] computer running bridge hangs when putting it under a high load

Wayback MachineAbout this captureCOLLECTED BY Organization: Alexa Crawls Starting in 1996, Alexa Internet has been donating their crawl data to the Internet Archive. Flowing in every day, these data are added to the Wayback Machine after an embargo period. Collection: Alexa Crawls DE Crawl data donated by Alexa Internet. This data is currently not publicly accessible TIMESTAMPSloadingLennert Buytenhekbuytenh@gnu.org
Wed, 23 Aug 2000 02:45:15 +0200 (CEST)Hi,Heh.. I believe I was involved in setting this bridge up :)It sounds like marginal hardware. Can you try to put it under severenetwork stress without using the bridge code? For example: assign each NICa different subnet and run flood pings against it. This should tell youwhether the bridge code is involved or not. I guess in this case, it’snot.greetings,LennertPS Almost certainly, the realtek cards have nothing to do with this.On Tue, 22 Aug 2000, Kristian Rietveld wrote:> Hi all!>> At school we’ve got a computer running linux-2.2.16br-0.0.6 which bridges> some segments of the BNC network. When we wanted to test the performance> of the bridge (to check if it’s faster than 18 computers on a single> 10Mbit BNC cable… 🙂 we started w95 (we only have 486 computers with 16> MB RAM, I know it’s slow … ) on all machines, they succeeded to find the> network and to log in. The clients access the server (NetWare 4) on the> UTP network through the bridge (which has 5 BNC connections and one UTP).> After a succeeded log on, we started Word or WordPerfect (all applications> are installed on the server) on all 18 clients at the same time. This is a> very high load on the bridge, after 23 seconds the bridging computer hangs> _completely_, we had to reset it (fortunately, this computer has an ext3> filesystem :).>> About the configuration of the computer, it currently runs> linux-2.2.16br-0.0.6. It’s a P200, with 16 MB RAM. It has the following> NICs:> – 3 3com Etherlink III (ISA) (we use the BNC connection of these cards)> – 1 NE2000 (ISA) (BNC)> – 1 NE2K-PCI (Realtek) (BNC)> – 1 3com Fast Etherlink (Cyclone, 100Mbit) (UTP)>> I heard about hangs with bridges using RealTek cards, in most cases only> the bridge hangs, and not the whole computer. So I don’t think this is the> cause of the problem. During the ‘performance test’, we only used 4> network cards (1 UTP (3com Cyclone), 3 BNC, don’t know which ones, so the> RealTek one may have been unused :). The other 2 cards weren’t used, the> clients attached to those 2 segments weren’t installed yet…>> So, summarizing, the bridge runs nearly perfectly until a very high load> is put on it.>> I hope somebody has some clues…>>> Kris>>

E2compr 0.4 User Manual – fragmented

Wayback MachineAbout this captureCOLLECTED BY Organization: Alexa Crawls Starting in 1996, Alexa Internet has been donating their crawl data to the Internet Archive. Flowing in every day, these data are added to the Wayback Machine after an embargo period. Collection: Alexa Crawls DE Crawl data donated by Alexa Internet. This data is currently not publicly accessible TIMESTAMPSloadingGo to the first, previous, next, last section, table of contents.

For over 99% of files, fragmentation is no worse than in an uncompressedfilesystem.(3) As the file isbeing written, we compress whenever we reach the end of a cluster. Whenwe start allocating blocks for the next cluster, we try to allocate themright next to the blocks of the previous, compressed, cluster. (Don’tworry, the ext2 block allocation strategy does the right thingconcerning holes.)

It is only when you write over a cluster other than the last in the filethat compression can cause extra fragmentation. This is because the newdata might compress to a different number of blocks than previously, soyou either get a gap in allocation (if the new data takes up fewerblocks) or you get a block that has to be written out of sequence (ifthe new data takes up more blocks than previously allocated to thecluster). The only example of this sort of file that I can think of islarge (more than one cluster) database files. (`updatedb’ doesn’tcount, because (I believe) it gets truncated to zero length before being written over.)

As someone else (kragen@pobox.com) pointed out, compression canactually reduce fragmentation in some cases, simply because we don’t fillup the disk as quickly.

Nevertheless, people who are short on disk space (as many e2compr usersare) tend to have a high turnover of files (deleting files to make wayfor new files, which have to be written in the cracks occupied by thefiles just deleted), which causes high fragmentation.See section Can I still use a defragmenter?, for comments on using a defragmenter.

Go to the first, previous, next, last section, table of contents.

e2compr: Mailing Lists

Wayback MachineAbout this captureCOLLECTED BY Organization: Alexa Crawls Starting in 1996, Alexa Internet has been donating their crawl data to the Internet Archive. Flowing in every day, these data are added to the Wayback Machine after an embargo period. Collection: Alexa Crawls DE Crawl data donated by Alexa Internet. This data is currently not publicly accessible TIMESTAMPSloading

List Descriptions

The e2compr list is for general discussion for both users and developers, concerning e2compr. Anyone can post to this list, whether as a new user asking for help, or to alert users and developers of a possible bug, or to suggest a new feature, or to propose an implementation of a feature and ask for comments. Announcements specific to e2compr-0.4 are sent to this list rather than to e2compr-announce.

The e2compr list is low-volume; it has in the order of one message per week.

The e2compr-announce list is currently reserved for announcements specific to e2compr-0.3 (i.e. old e2compr). Only Peter can post to this list.

It is strongly recommended that all e2compr users be subscribed to at least one of these lists (which depending on which version of e2compr you are using). That way, you’re kept up to date with bug fixes.

Thanks to Christoph Lameter for setting up these lists. Send any list problems to e2compr-admin@e2compr.memalpha.cx), though.

How to Subscribe / Unsubscribe

This is email command help for version 1.0rc2 of the “Mailman” list manager. The following describes commands you can send to get information about and control your subscription to Mailman lists at this site. A command can be in the subject line or in the body of the message.

Note that much of the following can also be accomplished via the World Wide Web, at http://e2compr.memalpha.cx/mailman/listinfo/e2compr and http://e2compr.memalpha.cx/mailman/listinfo/e2compr-announce (for the e2compr and e2compr-announce lists, respectively).

In particular, you can use the Web site to have your password sent to your delivery address.

List-specific commands (subscribe, who, etc.) should be sent to the *-request address for the particular list, i.e. e2compr-request@e2compr.memalpha.cx or e2compr-announce-request@e2compr.memalpha.cx for the e2compr and e2compr-announce lists respectively.

About the descriptions: Words in “<>“s signify REQUIRED items and words in “[]” denote OPTIONAL items. Do not include the “<>“s or “[]”s when you use the commands.

The following commands are valid:

subscribe [password] [digest-option] [address=

] Subscribe to the mailing list. Your password must be given to unsubscribe or change your options. When you subscribe to the list, you’ll be reminded of your password periodically. `digest-option’ may be either: `nodigest’ or `digest’ (no quotes!). If you wish to subscribe an address other than the address you send this request from, you may specify “address=” (no brackets around the email address, no quotes!). unsubscribe [address] Unsubscribe from the mailing list. Your password must match the one you gave when you subscribed. If you are trying to unsubscribe from a different address than the one you subscribed from, you may specify it in the `address’ field. whoSee everyone who is on this mailing list. info View the introductory information for this list. lists See what mailing lists are run by this Mailman server. help This message. set