Les stress socio-écologiques peuvent compromettre la reproduction des femelles et affecter le développement de leur progéniture. Par exemple, le stress gestationnel peut avoir de profonds effets sur la morphologie, le comportement, la physiologie et ultimement la valeur adaptative des jeunes. De tels effets maternels jouent un rôle particulièrement important chez les mammifères ayant un lien intime et prolongé avec leur progéniture. Les femelles se reproduisant avec l'aide coopérative d'un groupe doivent composer avec les tensions entre femelles reproductrices, qui mènent à la reproduction dans des contextes socio-écologiques plus ou moins stressants. Dans ce contexte, la thèse vise à identifier des indicateurs du stress gestationnel chez le suricate femelle (Suricata suricatta) en milieu naturel et investiguer les effets du stress maternel sur l'axe du stress de la progéniture. Le suricate est un mammifère grégaire coopératif chez qui le partage inégal de la reproduction entraîne une forte compétition entre les femelles, dont certaines se reproduisent hors des périodes de bonnes conditions alimentaires. Avant ce travail, le stress prénatal n'avait été étudié qu'en milieu contrôlé, entre autres faute d'une méthodologie adéquate permettant son estimation non invasive durant la reproduction de populations sauvages. Le suricate du Kalahari offrait alors une excellente opportunité d'étudier le lien entre le biais de la reproduction, les conditions socio-écologiques, le stress gestationnel et le phénotype hormonal de la progéniture. En effet, la disponibilité de données d'histoire de vie, de condition corporelle, de comportements sociaux et d'échantillons fécaux permettait l'étude longitudinale d'hormone de stress lors de la reproduction et les premiers moments de vie pour plus de 800 individus sauvages marqués depuis 1997. Le premier chapitre évalue l'effet de l'entreposage de longue durée d'échantillons fécaux sur les concentrations mesurées en hormones de stress (métabolites de glucocorticoïdes, fGC) et de reproduction (métabolites d'estrogènes, fE). Les résultats recommandent la lyophilisation précoce plutôt que l'entreposage d'échantillons frais à -20°C. De plus, l'étude souligne l'importance d'évaluer l'effet d'entreposage pour les différentes hormones étudiées et de contrôler pour le temps d'entreposage. Le deuxième chapitre étudie le lien entre le taux de reproduction, la dominance sociale et les fGC. Les changements en fGC au cours de la reproduction sont influencés par le chevauchement de reproductions, non par la dominance. L'étude remet ainsi en cause le rôle des GCs dans la suppression de la reproduction chez les subordonnées. Le troisième chapitre étudie les stress socio-écologiques avant et pendant la gestation et identifie la compétition entre femelles et le chevauchement de reproduction lors de mauvaises conditions alimentaires comme facteurs entraînant une augmentation en fGC. Le quatrième chapitre évalue l'effet du stress gestationnel sur l'axe du stress de la progéniture avant le sevrage. Les résultats mettent en évidence l'effet maternel du chevauchement des reproductions en période de stress alimentaire, entraînant une augmentation en fGC de la progéniture, et l'effet, dépendant du sexe de la progéniture, de la compétition entre les femelles sur les fGC de la progéniture après la naissance, entraînant une diminution en fGC chez les fils, non chez les filles. L'étude démontre ainsi l'utilité du suivi longitudinal des fGC pour l'étude des stratégies d'histoire de vie et des effets maternels hormonaux. Les effets du stress alimentaire sur les GCs maternels lors du développement prénatal suggèrent un conflit entre les stratégies permettant de maximiser le succès reproducteur maternel et les effets néfastes sur la santé future des jeunes. Finalement, les effets maternels observés suggèrent un potentiel d'action des GCs gestationnels sur la croissance et le phénotype comportemental de la progéniture. Par exemple, l'effet maternel du stress social dépendant du sexe de la progéniture pourrait aussi modifier leur potentiel physiologique à gérer les stress associés aux stratégies comportementales de coopération et reproduction.

In group living species animals commonly compete for limited resources such as food (Darwin 1859). Winning competition for food may be beneficial for an individuals survival or reproductive success (Williams 1966; Clutton-Brock 1988; Metcalfe et al 1995) but conflict with group members may be costly as it typically involves aggression (Huntingford&Turner 1987, Mesterton-Gibbons&Adams 1998). Asymmetries between individuals are predicted to determine the outcome of competition (Maynard-Smith&Parker 1976), and individuals are expected to steal food when the benefit to them is greatest (Barnard 1984; Trivers 1972). I therefore investigate what determines the outcome of competition for food between group members, and what factors affect whether group members try to steal food in the cooperatively breeding meerkat (Suricatta suricatta). Meerkats competed for food items infrequently and the owner of a food item typically won competition, but dominant individuals and breeding females were more likely to win competition than other group members. This provides support for models of conflict over resources in group living species which predict that ownership may determine the outcome of competition, thereby avoiding frequent costly conflict (Maynard-Smith 1982). Furthermore, where large asymmetries exist between contestants in dominance status or the value of a resource, these may determine the outcome of competition (Maynard-Smith&Parker 1976; Grafen 1987). Meerkats varied in how frequently they tried to steal food depending upon the costs and benefits of competition. Dominant individuals competed for food more frequently which is likely to reflect reduced costs of competition as subordinate individuals may avoid conflict with them (Packer&Pusey 1985). Females competed for food more frequently than males and more frequently during breeding, reflecting the higher costs of reproduction to females compared to males (Williams 1966; Trivers 1972). Meerkats compete more frequently for food when food availability is low, which indicates that food items may be more valuable when they are rare. Meerkats in smaller groups competed more frequently. In cooperatively breeding species group members undertake a large number of costly helping behaviours. Individuals in small groups each contribute more effort to helping than individuals in large groups and suffer higher costs which may increase the benefit of food to them (Clutton-Brock et al 1998a; Clutton-Brock et al 2001a). Competition for the opportunity to breed in cooperatively breeding meerkats has resulted in despotic dominance hierarchies where a dominant female monopolises breeding and reproductively suppresses subordinates (Clutton-Brock et al 2001b). Dominant females stole more food than any other group members. This is likely to be a consequence of the high costs of reproduction for the dominant breeding female in species with high reproductive skew (Creel&Creel 1991; Clutton-Brock et al 2001b). Furthermore, dominant females were more aggressive and more successful in competition for food with their reproductive competitors. Dominant females may therefore use competition for food as a means of asserting dominance over their reproductive competitors which could contribute to reproductive suppression (Creel et al 1992; Williams 2004; Kutsukake&Clutton-Brock 2006b; Young et al 2006). / Dissertation (MSc)--University of Pretoria, 2011. / Zoology and Entomology / unrestricted

This thesis deals with filtering selected DDoS attacks on denial of the service. The the toretical part deals with the problems of general mechanisms used for DDoS attacks, defense mechanisms and mechanisms of detection and filtration. The practical part deals with the filtering of attacks using the iptables and IPS Suricata firewall on the Linux operating system in an experimental workplace using a network traffic generator to verify its functionality and performance, including the statistical processing of output data from filter tools using the Elasticsearch database.

<p>Arbetets syfte är att jämföra intrångsdetekteringssystemen Snort och Suricata för att ge en uppfattning om vilken av applikationerna som lämpar sig att implementeras hos en internetleverantör för att upptäcka attacker och öka säkerheten på nätverket. Jämförelsen utförs med hänseende till antal upptäckta attacker, prestanda, implementeringstid, antal konfigurationsfiler samt vilka operativsystem de finns tillgängliga på.</p><p>Resultatet visar att Suricata med sitt stöd för att använda signaturer skapade för Snort upptäcker fler attacker än Snort. Snort däremot går både smidigare och snabbare att implementera. Prestandamässigt så visar Suricata bäst resultat, genom att använda sig av flera kärnor och mindre minne.</p>

Arbetets syfte är att jämföra intrångsdetekteringssystemen Snort och Suricata för att ge en uppfattning om vilken av applikationerna som lämpar sig att implementeras hos en internetleverantör för att upptäcka attacker och öka säkerheten på nätverket. Jämförelsen utförs med hänseende till antal upptäckta attacker, prestanda, implementeringstid, antal konfigurationsfiler samt vilka operativsystem de finns tillgängliga på. Resultatet visar att Suricata med sitt stöd för att använda signaturer skapade för Snort upptäcker fler attacker än Snort. Snort däremot går både smidigare och snabbare att implementera. Prestandamässigt så visar Suricata bäst resultat, genom att använda sig av flera kärnor och mindre minne.

Master's thesis is focused on intrusion detection for denied of service attacks. These distributed DoS attacks are threat for all users on the Internet, so there is deployment of intrusion detection and intrusion prevention systems against these attacks. The theoretical part describes the DoS attacks and its variants used most frequently. It also mentioned variants for detecting DoS attacks. There is also described, which tools are used to detect DDoS attacks most frequently. The practical part deals with the deployment of software tools for detecting DDoS attacks, and create traffic to test detection abilities of these tools.

Meerkat (Suricata suricatta) is a mongoose species from South Africa. Insects are the primary food sources of their diet. Meerkat lives in groups of 2?30 individuals with a developed social structure. The group consists of a dominant pair and a different number of helpers of both sexes, which are involved in various forms of cooperation, for example care for cubs, guard service and maintaince of burrow. Meerkats are at risk of predation by mammalian predators, avian predators and snakes. One meerkat always seems to be a sentry and stands guard to keep the gang safe. When the guard senses danger they give a warning bark and the other members run for cover into their burrow. This study was done from June to September 2011 in zoo in Jihlava on a group of meerkats, which consisted of 13 individuals. There was used the method of direct observation by a scanning behaviour (behaviour sampling). Observation was specialized on sentinel behaviour. The main aim of this work was evaluate the results of observation and then compare this results with behaviour in the wild. By monitoring was demonstrably found, that dominant individuals contributed to guarding more frequently and the longest time and the juveniles contributed at least. There wasn´t set order in guarding. The most often was individually guarding or guarding in a group of 2 or 3 individuals.

Meerkats are known to strongly rely on chemical communication in social contexts. However, little is known about their use of the sense of smell in food detection and selection. The aim of the present study was therefore to assess whether meerkats are able to (1) detect hidden food using olfactory cues, (2) distinguish the odour of real food from a single food odour component, and (3) build an association between the odour of real food and a novel odour. I employed the buried food test, widely used with rodents to assess basic olfactory abilities, designed to take advantage of the propensity of meerkats to dig. I found that the meerkats were clearly able to find all four food types tested (mouse, chicken, mealworm, banana) using olfactory cues alone and that they successfully discriminated between the odour of real food and a food odour component. In both tasks, the animals dug in the food-bearing corner of the test arena as the first one significantly more often than in the other three corners, suggesting development of an efficient foraging strategy. No significant association-building between a food odour and a novel odour was found within the 60 trials performed per animal. I conclude that meerkats are able to use olfactory cues when foraging and that their sense of smell is well-adapted for recognizing specific odours of behavioural relevance. To the best of my knowledge, this is the first study to successfully employ the buried food test with a carnivore species.

Context. The intrusion detection systems are being widely used for detecting the malicious traffic in many industries and they use a variety of technologies. Each IDs had different architecture and are deployed for detecting malicious activity. Intrusion detection system has a different set of rules which can defined based on requirement. Therefore, choosing intrusion detection system for and the appropriate environment is not an easy task. Objectives. The goal of this research is to evaluate three most used open source intrusion detection systems in terms of performance. And we give details about different types of attacks that can be detected using intrusion detection system. The tools that we select are Snort, Suricata, OSSEC. Methods. The experiment is conducted using TCP, SCAN, ICMP, FTP attack. Each experiment was run in different traffic rates under normal and malicious traffics all rule are active. All these tests are conducted in a virtual environment. Results. We can calculate the performance of IDS by using CPU usage, memory usage, packet loss and a number of alerts generated. These results are calculated for both normal and malicious traffic. Conclusions. We conclude that results vary in different IDS for different traffic rates. Specially snort showed better performance in alerts identification and OSSEC in the performance of IDS. These results indicated that alerts are low when the traffic rates high are which indicates this is due to the packet loss. Overall OSSEC provides better performance. And Snort provides better performance and accuracy for alert detection.

The semestral thesis deals with the analysis of the modern open-source NIDPS tools for monitoring and analyzing the network traffic. The work rates these instruments in terms of their network location and functions. Also refers about more detailed analysis of detecting and alerting mechanisms. Further analyzes the possibilities of detection of anomalies, especially in terms of statistical analysis and shows the basics of other approaches, such as approaches based on data mining and machine learning. The last section presents specific open-source tools, deals with comparison of their activities and the proposal allowing monitoring and traffic analysis, classification, detection of anomalies and (D)DoS attacks.