Mercurial > dropbear
view libtomcrypt/demos/hashsum.c @ 1790:42745af83b7d
Introduce extra delay before closing unauthenticated sessions
To make it harder for attackers, introduce a delay to keep an
unauthenticated session open a bit longer, thus blocking a connection
slot until after the delay.
Without this, while there is a limit on the amount of attempts an attacker
can make at the same time (MAX_UNAUTH_PER_IP), the time taken by dropbear to
handle one attempt is still short and thus for each of the allowed parallel
attempts many attempts can be chained one after the other. The attempt rate
is then:
"MAX_UNAUTH_PER_IP / <process time of one attempt>".
With the delay, this rate becomes:
"MAX_UNAUTH_PER_IP / UNAUTH_CLOSE_DELAY".
author | Thomas De Schampheleire <thomas.de_schampheleire@nokia.com> |
---|---|
date | Wed, 15 Feb 2017 13:53:04 +0100 |
parents | 6dba84798cd5 |
children |
line wrap: on
line source
/* LibTomCrypt, modular cryptographic library -- Tom St Denis * * LibTomCrypt is a library that provides various cryptographic * algorithms in a highly modular and flexible manner. * * The library is free for all purposes without any express * guarantee it works. */ /* * Written by Daniel Richards <[email protected]> 6/7/2002 * hash.c: This app uses libtomcrypt to hash either stdin or a file * This file is Public Domain. No rights are reserved. * Compile with 'gcc hashsum.c -o hashsum -ltomcrypt' * This example isn't really big enough to warrent splitting into * more functions ;) */ #include <tomcrypt.h> #if defined(_POSIX_C_SOURCE) && _POSIX_C_SOURCE >= 200112L #include <libgen.h> #else #define basename(x) x #endif #if !defined(PATH_MAX) && defined(_MSC_VER) #include <windows.h> #define PATH_MAX MAX_PATH #endif /* thanks http://stackoverflow.com/a/8198009 */ #define _base(x) ((x >= '0' && x <= '9') ? '0' : \ (x >= 'a' && x <= 'f') ? 'a' - 10 : \ (x >= 'A' && x <= 'F') ? 'A' - 10 : \ '\255') #define HEXOF(x) (x - _base(x)) static char* hashsum; static void cleanup(void) { free(hashsum); } static void die(int status) { unsigned long w, x; FILE* o = status == EXIT_SUCCESS ? stdout : stderr; fprintf(o, "usage: %s -a algorithm [-c] [file...]\n\n", hashsum); fprintf(o, "\t-c\tCheck the hash(es) of the file(s) written in [file].\n"); fprintf(o, "\t\t(-a not required)\n"); fprintf(o, "\nAlgorithms:\n\t"); w = 0; for (x = 0; hash_descriptor[x].name != NULL; x++) { w += fprintf(o, "%-14s", hash_descriptor[x].name); if (w >= 70) { fprintf(o, "\n\t"); w = 0; } } if (w != 0) fprintf(o, "\n"); exit(status); } static void printf_hex(unsigned char* hash_buffer, unsigned long w) { unsigned long x; for (x = 0; x < w; x++) { printf("%02x",hash_buffer[x]); } } static void check_file(int argn, int argc, char **argv) { int err, failed, invalid; unsigned char is_buffer[MAXBLOCKSIZE], should_buffer[MAXBLOCKSIZE]; char buf[PATH_MAX + (MAXBLOCKSIZE * 3)]; /* iterate through all files */ while(argn < argc) { char* s; FILE* f = fopen(argv[argn], "rb"); if(f == NULL) { int n = snprintf(buf, sizeof(buf), "%s: %s", hashsum, argv[argn]); if (n > 0 && n < (int)sizeof(buf)) perror(buf); else perror(argv[argn]); exit(EXIT_FAILURE); } failed = 0; invalid = 0; /* read the file line by line */ while((s = fgets(buf, sizeof(buf), f)) != NULL) { int tries, n; unsigned long hash_len, w, x; char* space = strstr(s, " "); /* skip lines with comments */ if (buf[0] == '#') continue; if (space == NULL) { fprintf(stderr, "%s: no properly formatted checksum lines found\n", hashsum); goto ERR; } hash_len = space - s; hash_len /= 2; if (hash_len > sizeof(should_buffer)) { fprintf(stderr, "%s: hash too long\n", hashsum); goto ERR; } /* convert the hex-string back to binary */ for (x = 0; x < hash_len; ++x) { should_buffer[x] = HEXOF(s[x*2]) << 4 | HEXOF(s[x*2 + 1]); } space++; if (*space != '*') { fprintf(stderr, "%s: unsupported input mode '%c'\n", hashsum, *space); goto ERR; } space++; for (n = 0; n < (buf + sizeof(buf)) - space; ++n) { if(iscntrl((int)space[n])) { space[n] = '\0'; break; } } /* try all hash algorithms that have the appropriate hash size */ tries = 0; for (x = 0; hash_descriptor[x].name != NULL; ++x) { if (hash_descriptor[x].hashsize == hash_len) { tries++; w = sizeof(is_buffer); if ((err = hash_file(x, space, is_buffer, &w)) != CRYPT_OK) { fprintf(stderr, "%s: File hash error: %s: %s\n", hashsum, space, error_to_string(err)); ERR: fclose(f); exit(EXIT_FAILURE); } if(XMEMCMP(should_buffer, is_buffer, w) == 0) { printf("%s: OK\n", space); break; } } } /* for */ if (hash_descriptor[x].name == NULL) { if(tries > 0) { printf("%s: FAILED\n", space); failed++; } else { invalid++; } } } /* while */ fclose(f); if(invalid) { fprintf(stderr, "%s: WARNING: %d %s is improperly formatted\n", hashsum, invalid, invalid > 1?"lines":"line"); } if(failed) { fprintf(stderr, "%s: WARNING: %d computed %s did NOT match\n", hashsum, failed, failed > 1?"checksums":"checksum"); } argn++; } exit(EXIT_SUCCESS); } int main(int argc, char **argv) { int idxs[TAB_SIZE], idx, check, y, z, err, argn; unsigned long w, x; unsigned char hash_buffer[MAXBLOCKSIZE]; hashsum = strdup(basename(argv[0])); atexit(cleanup); /* You need to register algorithms before using them */ register_all_ciphers(); register_all_hashes(); if (argc > 1 && (strcmp("-h", argv[1]) == 0 || strcmp("--help", argv[1]) == 0)) { die(EXIT_SUCCESS); } if (argc < 3) { die(EXIT_FAILURE); } for (x = 0; x < sizeof(idxs)/sizeof(idxs[0]); ++x) { idxs[x] = -2; } argn = 1; check = 0; idx = 0; while(argn < argc){ if(strcmp("-a", argv[argn]) == 0) { argn++; if(argn < argc) { idxs[idx] = find_hash(argv[argn]); if (idxs[idx] == -1) { struct { const char* is; const char* should; } shasum_compat[] = { #ifdef LTC_SHA1 { "1", sha1_desc.name }, #endif #ifdef LTC_SHA224 { "224", sha224_desc.name }, #endif #ifdef LTC_SHA256 { "256", sha256_desc.name }, #endif #ifdef LTC_SHA384 { "384", sha384_desc.name }, #endif #ifdef LTC_SHA512 { "512", sha512_desc.name }, #endif #ifdef LTC_SHA512_224 { "512224", sha512_224_desc.name }, #endif #ifdef LTC_SHA512_256 { "512256", sha512_256_desc.name }, #endif { NULL, NULL } }; for (x = 0; shasum_compat[x].is != NULL; ++x) { if(XSTRCMP(shasum_compat[x].is, argv[argn]) == 0) { idxs[idx] = find_hash(shasum_compat[x].should); break; } } } if (idxs[idx] == -1) { fprintf(stderr, "%s: Unrecognized algorithm\n", hashsum); die(EXIT_FAILURE); } idx++; if ((size_t)idx >= sizeof(idxs)/sizeof(idxs[0])) { fprintf(stderr, "%s: Too many '-a' options chosen\n", hashsum); die(EXIT_FAILURE); } argn++; continue; } else { die(EXIT_FAILURE); } } if(strcmp("-c", argv[argn]) == 0) { check = 1; argn++; continue; } break; } if (check == 1) { check_file(argn, argc, argv); } if (argc == argn) { w = sizeof(hash_buffer); if ((err = hash_filehandle(idxs[0], stdin, hash_buffer, &w)) != CRYPT_OK) { fprintf(stderr, "%s: File hash error: %s\n", hashsum, error_to_string(err)); return EXIT_FAILURE; } else { for (x = 0; x < w; x++) { printf("%02x",hash_buffer[x]); } printf(" *-\n"); } } else { for (z = argn; z < argc; z++) { for (y = 0; y < idx; ++y) { w = sizeof(hash_buffer); if ((err = hash_file(idxs[y],argv[z],hash_buffer,&w)) != CRYPT_OK) { fprintf(stderr, "%s: File hash error: %s\n", hashsum, error_to_string(err)); return EXIT_FAILURE; } else { printf_hex(hash_buffer, w); printf(" *%s\n", argv[z]); } } } } return EXIT_SUCCESS; } /* ref: $Format:%D$ */ /* git commit: $Format:%H$ */ /* commit time: $Format:%ai$ */